<%BANNER%>

An Exploratory data collection approach for the assessment of level of service from a traveler's perspective

University of Florida Institutional Repository

PAGE 1

AN EXPLORATORY DATA COLLECTION APPROACH FOR THE ASSESSMENT OF LEVEL OF SERVICE FROM A TRAVELERS PERSPECTIVE By KIMBERLY SEAGER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Kimberly Seager

PAGE 3

This document is dedicated to my two grandfathers who have passed away since I began college.

PAGE 4

ACKNOWLEDGMENTS I would like to thank my committee chair and my committee chair members, Dr. Washburn, Professor Courage, and Dr. Li. I would also like to thank Cody for his help with all the equipment. Id like to thank Brad Choi for going with me to collect data and for convincing me to do a thesis. I would also like to thank David Kirshner and Luke McLeod for helping me on this project. iv

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS ............................................................................................... iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT.......................................................................................................................ix CHAPTER 1 INTRODUCTION......................................................................................................1 General Background...................................................................................................1 Problem Statement.....................................................................................................2 Current Methodology Used to Define Service Measures......................................2 Applicable Service Measures for Urban and Rural Freeways..............................3 Research Objectives and Tasks..................................................................................5 Organization of the Remainder of the Document......................................................5 2 LITERATURE REVIEW...........................................................................................6 Studies Pertaining to User Perceived LOS.................................................................6 Background Survey..................................................................................................17 Conclusions ..............................................................................................................18 HCM Freeway Analysis Methodology Overview....................................................18 Urban and Rural Freeway Overview...................................................................18 Applicable Thresholds for Urban and Rural Freeways.......................................19 3 RESEARCH APPROACH.......................................................................................20 Alternative Approaches Evaluation.........................................................................20 Video Data Collection Method................................................................................22 Vehicle Instrumentation......................................................................................22 Data Collection Site Selection.............................................................................24 Video Data Collection Runs................................................................................27 Video Clip Composition......................................................................................32 Loop Data Collection..........................................................................................33 v

PAGE 6

Pilot Test.............................................................................................................34 Pilot Test Procedure.......................................................................................35 Survey.............................................................................................................35 4 RESULTS.................................................................................................................38 Pilot Test Participant Feedback on Video Clips.......................................................38 Pilot Test Participant Feedback on Survey Instrument............................................39 Factors That Were Important in Evaluating Quality of Service...............................39 A Comparison of the HCMs LOS Evaluation to the LOS Ratings From the Pilot Test.......................................................................................................40 5 CONCLUSIONS AND RECOMMENDATIONS...................................................43 Conclusions ..............................................................................................................43 Vehicle Instrumentation......................................................................................43 Determination of Data Collection Sites..............................................................43 Collecting Video Data.........................................................................................44 Composition of Video Clips................................................................................44 Data Collection....................................................................................................44 Pilot Test.............................................................................................................45 Recommendations....................................................................................................45 APPENDIX A COMMON ABBREVIATIONS..............................................................................46 B POSSIBLE SITE LOCATION TRAFFIC INFORMATION..................................47 C FINAL SITE LOCATION DETAILS......................................................................52 D SITE LOCATIONS AND DRIVING TIME...........................................................59 E LOOP DATA SUMMARY TABLES......................................................................63 F PILOT TEST SURVEY...........................................................................................71 LIST OF REFERENCES...................................................................................................75 BIOGRAPHICAL SKETCH.............................................................................................77 vi

PAGE 7

LIST OF TABLES Table page 3-1. Study Approach Alternatives Matrix.......................................................................23 3-2. Final Site Location Information Table.....................................................................30 3-3. Loop Detector Traffic Information..........................................................................36 4-1. Pilot Test Participants Factors Affecting Trip Quality.............................................39 4-2. Comparison of Results.............................................................................................42 vii

PAGE 8

LIST OF FIGURES Figure page 3-1. Test Vehicle..............................................................................................................24 3-2. Front-View Camera Set-up......................................................................................25 3-3. Side-View Camera Picture.......................................................................................25 3-4. Speedometer Camera Picture...................................................................................25 3-5. Equipment Connection Schematic...........................................................................25 3-6. Possible Site Locations.............................................................................................28 3-7. Final Site Location Map...........................................................................................29 3-8. Composite Video Screenshot...................................................................................33 viii

PAGE 9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science AN EXPLORATORY DATA COLLECTION APPROACH FOR THE ASSESSMENT OF LEVEL OF SERVICE FROM A TRAVELERS PERSPECTIVE By Kimberly Seager May 2004 Chair: Scott Washburn Major Department: Civil and Coastal Engineering Although the concept of level of service (LOS) for freeways is usually defined in terms of users perceptions, there have been very few studies that have sought drivers or passengers views about what is important to them. The objective of this study was to develop and test a method by which driver perceptions of LOS for varying roadway conditions can be collected. The approach decided upon was to create video clips of rural freeway driving conditions from the drivers perspective. These video clips consisted of the front-view, the rear-view, the side-view, and the speedometer. Each video clip was between 2 to 2 minutes. In total, there were nine video clips. These video clips were then shown to seven pilot test participants. The participants were asked to rank the quality of their trip, the factors that influenced their trip quality, and the effectiveness of the video clips in terms of their ability to give the participants a feel for the traffic and roadway conditions they would experience if they were actually driving. An open ix

PAGE 10

conversation was then held for the researchers to get advice about how to improve the videotapes. The four most common factors that influenced the participants trip quality were traffic volume, pavement quality, ability to maintain a constant speed, and the percentage of trucks. The participants results were then compared to the density-based LOS calculated by the Highway Capacity Manual (HCM). Out of the nine video clips, three were rated the same as the HCMs density-based LOS. Five other clips were rated one letter grade lower than the HCMs density-based LOS ranking. The last video clip was rated three letter grades lower than the HCMs density-based LOS ranking. Some suggestions and recommendations given by the pilot test participants were to add sound, improve the video quality, videotape during different times of the day and varying weather conditions, and videotape more peak hour traffic conditions. Two complaints about this method were that it was hard to determine the roadways pavement quality and it was hard for some participants to compare two-lane rural freeways to three-lane rural freeways. Out of the seven participants, three ranked this videotape method as very good. Another three ranked this method as good. One participant ranked this method as fair. x

PAGE 11

CHAPTER 1 INTRODUCTION General Background Transportation engineers are continually faced with identifying roadway infrastructure improvements that will result in good benefit-to-cost ratios. Due to limited financial resources, it is extremely important to make good decisions regarding selected improvements. One of the most commonly used tools in the decision-making process is the Highway Capacity Manual (HCM). The HCM is the basic reference and procedural guide for traffic operations of analyses in the United States. It is one of the major resources used to predict operational improvements resulting from a new or improved facility. The HCM provides a systematic and consistent basis for analyzing the capacity and level of service (LOS) for various highway and street facilities, including signalized and unsignalized intersections, arterial streets, freeway segments, and highway segments. The HCM is published by the Transportation Research Board (TRB) of the National Academy of Sciences and is developed by the Highway Capacity and Quality of Service (HCQS) committee. A new edition of the HCM was published in 2000, the first complete revision of the HCM since 1985. LOS assessment of a roadway facility has become a major foundation of the HCM. The LOS concept is used in the HCM as a qualitative indicator of a travelers trip quality under specified roadway, traffic, and control conditions. In the 2000 version of the HCM, LOS is described as a qualitative measure describing operational conditions within 1

PAGE 12

2 a traffic stream, based on service measures such as speed and travel time, freedom to maneuver, traffic interruptions, comfort, and convenience [1]. The 2000 HCM divides quality of service into six levels of service, LOS A LOS F. LOS A represents excellent service and LO S F represents very poor service. If the HCM analysis yields LOS A, the user knows clearly that the road is uncongested. This facility has very low traffic volume and vi rtually no traffic conge stion. At LOS C the facility is in the mid-range of congestion. If the facility is at LOS E, it is approaching its capacity, but still has undersa turated conditions. LOS F represents oversaturated conditions. This occurs when the traffic demand exceeds the capacity of the facility. Thus, the capacity of a roadway or intersec tion, as defined in the HCM, represents the boundary between LOS E and F. Problem Statement Current Methodology Used to Define Service Measures There is no commonly accepted quantitative procedure for setting LOS threshold values (the service measure values that de lineate one LOS value from another). The criteria and breakpoints used to define LOS are based on the collective professional judgment of the members of the HCQS comm ittee of the TRB. Therefore, the LOS threshold values represent the perspectives of transportation experts. The service measure for each facility type has generally been selected based on two considerations. The first consideration is that the service measure chosen should represent speed and travel time, freedom to maneuver, traffic in terruptions, and comfort and convenience in a manner most appropriate to char acterizing quality of service for the particular facility type being analyzed. The second consideration is that the service measure chosen should

PAGE 13

3 be sensitive to traffic flow rates so that th e service measure charac terizes the degree of congestion of the facility [2]. However, the 1985 HCM described LOS as a qualitative measure that characterizes operational conditions within a traffic stre am and their perception by motorists and passengers. The descriptions of individual levels of servi ce characterize th ese conditions in terms of such factors as speed and tr avel time, freedom to maneuver, traffic interruptions, and comfort and convenience [3]. The quotation states that the perception of driving conditions by drivers and passengers is an essential element in evaluating the quality of service on a freeway facility. In addition, little has been done to compare the expert opinions of the HCQS committee with the users perceptions of LOS despite a definition that suggests that it is driven by or meant to reflect user perception. When these service measures were chosen the committee probably felt that they were highly correlated with user perceptions, but they did not know for sure since users have rarely been asked how they felt. Research on traveler perceptions of quality of service is desirable so that the existing service measures can be validated (tha t the existing service measures and thresholds are reflective of how roadway user s actually perceive conditions). This will assure transportati on engineers and infrastructure investment decision-makers that improvements identified as a result of LOS analyses will actually be perceived as improvements by the traveling public. This will give the traveling public more faith in the abilities of the trans portation engineering community and those in charge of managing the transportation system. Applicable Service Measures fo r Urban and Rural Freeways Rural freeways differ from urban freeways in that interchange spacing is much greater for rural freeways, posted speed limits are higher for rural freeways, and rural

PAGE 14

4 freeways have a much larger percentage of social and recreational trips while urban freeways have a higher percentage of shoppi ng and work trips, to name a few main differences. Yet rural and urban freeways ha ve the same service measure (density) and thresholds for LOS, even though traveler perceptions and expect ations are probably different for rural freeway trip s and urban freeway trips. This belief is based on the premise that while urban freeway travelers experience the full range of LOS values, rural freeway tr avelers rarely experience density conditions that by current definition exceed LOS C. It is believed that travelers on rural freeways usually expect low-density conditions, and even moderate amounts of congestion can have a negative impact on a travelers pe rceived quality of service being experienced [4]. If driver perceptions and expectations diffe r between rural and urban areas, the question that arises is whether rural and urban fr eeways should have the same service measure and/or thresholds. The concept of different service measures and/or thresholds for a certain facility type in areas with differing levels of devel opment is nothing new. In fact, it is already being used for a couple of facility types. Other roadway facility analysis methodologies of the HCM have divided the faci lity type into different clas ses. For example, two-lane highways have been divided into two classes. The two-lane highway analysis procedure currently defines two classifi cations of highway by primar y trip type served, with differing service measures and thresholds. For the service measure in common between the two classes (percent time spent following), the thresholds are different. Additionally, the urban arterials analysis methodology uses one service measure, average travel speed,

PAGE 15

5 but four different sets of thresholds, corre sponding to four differe nt arterial street classifications. Research Objectives and Tasks The objective of this stu dy was to develop and test a method by which driver perceptions of LOS for varyi ng roadway conditions can be collected. The method that was chosen for this study was an in-vehicle video that would be shot from a drivers perspective. This will be described in detail in the research approach chapter. It is hoped that this method will provide the means to develop a procedure for estimating LOS on rural freeways based on drivers perceptions. Five major tasks were performed in the development of this method. These tasks are as follows: Determine how best to instrument the dr iving vehicle for the collection of rural freeway data via video Determine the rural freeway sites from which to collect video footage and traffic operations data Coordinate the simultaneous collection of loop detector data for the chosen sites Produce the final video clips to be us ed for survey participant review Pilot test the video scen es and LOS ranking survey Organization of the Remainder of the Document There are four remaining chap ters in this document. Chapter 2 contains a summary of relevant literature and an overview of the 2000 HCM freeway analysis methodology. Chapter 3 contains the details of the resear ch approach used for this study. Chapter 4 describes the results from the pilot test. Th e conclusions and recommendations section is contained in Chapter 5

PAGE 16

CHAPTER 2 LITERATURE REVIEW Although the HCM states that the concept of LOS should reflect traveler perception there have been very few studies that have sought travelers views about what factors are important to them in assessing trip quality. A literature review was conducted to identify previous research efforts into the specific issue of LOS perceptions by roadway users. Studies Pertaining to User Perceived LOS A study by Pcheux et al. [5] discusses how the right performance measures are not being used to measure traveler perceptions and satisfaction. The authors felt this way because there have not been many studies of drivers perceptions done or written about. Because there have been so few studies conducted to date in the area of traveler perception of quality of service, it is safe to assume that the performance measures currently in the HCM probably do not represent those most important to travelers. Pcheux et al. feel that if the HCM really means that LOS is meant to reflect drivers and passengers perceptions of the quality of service on a highway, then engineers cannot choose the performance measures as they have in the past. Instead, the public needs to decide what performance measures should be used to determine quality of service. This paper concluded that more studies need to be done focusing on traveler perceptions. A study by Sutaria and Haynes [6] used a road-user opinion survey to evaluate the current LOS methodology at signalized intersections. The survey involved depicting and rating different traffic situations at a signalized intersection. Two types of film sequences 6

PAGE 17

7 were used: a drivers view and an overhead view of an intersection. Over 300 drivers rated randomly arranged films in terms of appropriate LOS. Field studies and the attitude survey given to the drivers su rveyed provided data for the de velopment of two models. According to the 1965 HCM, LOS is a qua litative measure of the effect of a number of factors, which include speed, tr avel time, traffic interruption, freedom to maneuver, safety, driving comfort, convenience, and operating costs [7]. Since there are stops at intersections, speed cannot be the appr opriate measure of LOS for arterials. That led to the HCM establishing an operational index called load factor (LF). The HCM defined this index as a ratio of the total number of green si gnal intervals that are fully utilized by traffic during the p eak hour to the total number of green intervals, for that approach during the same period [7]. The LF was then used to determine the various LOS at signalized intersections. Sutaria and Haynes decided that their premis e would be that the quality of flow at any intersection should reflect the attitudes of the road users. Since there are different levels of satisfaction regardi ng intersection op eration, they felt that the opinions of a group of road users could be used to esta blish a rational relationship between LF and LOS. To do this would involve a compilation of all the road users inclinations, feelings, and degrees of satisfaction about the quality of service at an inte rsection. Drivers subjective ratings of quality of flow at a signalized intersection would re present their attitudes or opinions, and LF, average individual delay (AID ), or a saturation factor would represent an objective rating. The film segments chosen would represen t various specific leve ls of service and would be shown to the drivers for a duration of one or two signal cycles. The final film

PAGE 18

8 prepared was a compellation from one or two representative film segments for each LOS. The final film also included both types of vi ew, and the order of th e segments in each type was determined on a random basis. The survey consisted of two parts: a group attitude survey and the film survey. A user-response survey was decided upon because such an approach had been successfully utilized previously and showed that it coul d provide meaningful results. A questionnaire was developed for the group attitude survey. Dr ivers were asked to answer the first part of the questionnaire before they viewed the se gments of film. The first part inquired about sex, age, driving experien ce, education, and type of road way most often driven on. They were then requested to rate their opini ons of five factors re garding the quality of service at an intersec tion: delay, number of stops, traffic congestion, number of trucks and buses, and difficulty of lane changing. Then the first part of the film was presented to them segment by segment. Each segment c onsisted essentially of a drivers view of approaching, waiting, and passing through the intersection. After each segment the group was requested to rate the traffic operations with regard to LOS provided. After the presentation of the film, the dr ivers were requested to again ra te the five factors regarding quality of flow at the intersection. This was done in order to find out if the films influenced their initial opinions. In all, 310 respondents participated in the survey at several sessions. The results from the group attitude survey indicated that delay was considered the most important factor both before and after viewing the film segments. The other four factors changed in rank before and after viewi ng the films. Based on the results from the film survey, the hypothesis that LF is a better predictor of LOS than AID was not true.

PAGE 19

9 This strongly supports the argument that the definition of LOS as a function of LF was not adequate. This lead to a new relati onship among AID, LF, volume to capacity (v/c) ratio, and LOS based on user perception. This st udy was one of the first that asked users how they felt about traffic conditions. The results from this study changed the measure used by the HCM to assess LOS at signalized intersections. An additional study conducted by Pche ux et al. [8] addresses methodological issues faced in the development of a study to assess user perceived LOS at signalized intersections. This study had two main objectives. The first objective was to determine the appropriateness of the current LOS methodology by seekin g users opinions. The second objective was to determine the factors affecting users LOS perceptions at signalized intersections. Although the concept of LOS is meant in pa rt to reflect the operational conditions as perceived by motorists, the HCM LOS thres holds for signalized in tersections were not based on studies of driver perceptions. Th e HCM criteria were created from observed field delays. Thus, each LOS represents diffe rent delay conditions, but not necessarily the delay that motorists perceived. The HC M specifies average control delay as the measure of effectiveness for signalized intersection LOS analysis, but it is unlikely that delay is the only factor that influences user perception of quality of service. The HCM uses the same delay thresholds that were established 15 years ago. Thus, travel conditions that would have been viewed as intolerable in the 1960s are considered normal by todays motorists, es pecially commuters. Methods of data collection evaluated for this project included: on-the-road field studies, controlled test-track studies, and controlled laborat ory studies. The researchers

PAGE 20

10 decided that subjects would be shown videos as though they were the driver. This method would allow for multiple subjects to be run simultaneously and give the researchers control over the experi mental conditions. A questionnaire with three parts was designed to gather detailed subject information. The first part was designed to explore the subjects attitudes about driving in certain situations. The second part of the questionnaire was designed to explore personal characteristics of the subjects. So cio-demographic information was obtained in the third part of the questionnaire. Subjects were not persuaded to use delay for rating LOS, they were simply told to rate the quali ty of service provided by the traffic signal. The results of the survey show that subjects were more tolerant of delays than the HCM would suggest, and their LOS ratings tend ed to be similar for delays associated with LOS A and B and LOS C and D. The results also showed that subjects delay estimates were fairly accurate, but widely va riable, as were their perceptions of LOS. Although a few subjects used delay as the only criterion when rating LO S, most subjects considered more than just delay. Fifteen factors were identified by the subjects as influential in their LOS ratings. They were as follows: delay size of intersection traffic signal efficiency pavement quality arrows/lanes for turning vehicles queue length visibility of traffic signals from queue traffic mix clear/legible signs and road markings location geometric design of intersection scenery/aesthetics leading left-turn phasing scheme presence of pedestrians

PAGE 21

11 visual clutter/distractions These results suggest that users percepti ons of LOS are sensitive to factors other than delay. The authors suggest in future experiments that the mo st important factors influencing users perceptions fi rst be identified and then controlled in the experimental design. A study by Hall et al. [9] reports on the re sults of focus group sessions in which a group of commuters discussed their views about determinants of the freeway quality of service they experienced. The researchers tried to figure out what aspects of freeway travel are important to motorists. Hall et al. concluded that total travel time is the most important determinant for commuters, but a number of other aspects of the trip also mattered. These aspects included safety, tr aveler information, and maneuverability (density). This research used focus groups in order to determine what aspects of freeway travel are most important to motorists. There were two focus groups, one with five people, and the other with seven members. Focus group members were all university faculty members, in a wide range of depa rtments. The total population of drivers included non-commuters as well as commuters truck drivers as well as drivers of passenger cars, and a wide range of backgr ounds and education. More men than women participated in this study. Thes e people were selected to ensure that all of the participants traveled the same stretch of freeway, so th at they all knew about the situations being discussed, and had relatively similar experiences The research team felt that this would provide the most productive context for an expl oratory analysis of us ers perceptions.

PAGE 22

12 The focus groups lasted an average of one and a half hours. In the focus groups, perceptions of trip quality and the factors that influence trip quality were investigated, as well as the factors that influence changes in pe rception from trip to trip. The facilitator guided the focus group participants using an in terview topic checklist containing a list of the topics relevant to the research, and a se ries of semi-structure d, open-ended questions designed to draw out information on these t opics. Finally, the re searchers asked about the differences between being a driver and be ing a passenger. This was included in the topics discussed because the HCM states that the perceptions of both drivers and passengers are important. Preliminary results we re circulated to research participants to help assess the credibility of th e researchers interpretations of the experiences discussed. Responses indicate that the focus group part icipants agree with the researchers interpretations of their discussions. The results from the focus groups identif ied four primary factors that were important to urban freeway motorists. These factors were travel time, density, safety, and traveler information. Travel time is the first thing respondents used to describe the quality of a particular trip. How long it take s to get from where th ey are to where they need to be was the most important measure of the quality of that par ticular trip. Having time constraints on their arrival greatly increases the stress involved in the trip and with it the perceived quality of the tri p. The participants also felt time spent commuting was lost time as far as they were concerned. An indi cator of the importance of travel time was that when researchers asked about trips that were less than desirable the first item that came up in response was construction. Constr uction was mentioned because it tended to slow traffic down. The other indicator was the length some people went to in order to

PAGE 23

13 keep the travel time relatively short. An example of this would be leaving the house for work at 4 a.m. instead of 6 a.m. A number of other issues were mentione d in the focus group: density, accidents, travel information, driver civility (or incivil ity), the use of photo radar, and weather. The way the focus groups felt about density was best summarized by the following comment. Even if theres an extraordinar y volume of traffic, as long as its clicking along, even if theres cars all around you, and trucks and ev erything, as long as everybodys going then its just a question of time [9 ]. Participants were concer ned about accidents not only in terms of congestion, but also because of th e risk to their pers onal safety. Having adequate information about what was happeni ng to traffic while they were on the road was also important to participants. Driver civility and politeness were mentioned in reference to the notion of lane etiquette (i.e. using different lanes for different speeds and to pass). The responses of both focus groups made it clear that people have a very different perspective on the trip when they are passengers as compared to drivers. The focus group participants felt that the tension and sense of responsibility are much less as a passenger than as a driver. Also, there is not the same degree of concern about the travel time. This paper was helpful because focus groups were one of the options considered in the alternative approaches evaluation. This paper also showed that density is not the only factor considered by drivers. A paper by Hostovsky and Hall [10] deals wi th the perceptions of tractor-trailer drivers regarding the quality of service on freeways. This was accomplished by a focus group with professional tractor-t railer drivers. Because of the different performance

PAGE 24

14 characteristics between automobiles and trac tor trailers, their reasons for being on the road, and the amount of time spent on the road, it is reasonable to expect that different factors will affect truck driver s perceptions of freeway quality of service. The paper also compares and contrasts tractor-trailer drivers perceptions with factor s that are important to freeway automotive commuters. The focus group was held on November 15, 2001 and consisted of seven tractortrailer drivers. The sessi on lasted for one and a half hours. The focus group was structured the same way as de scribed in Halls previous paper [5]. The researchers concluded that freeway conditions in general were the most frequently mentioned factors. The three variables that were used to describe traffic conditions were: travel time (or speed), traffic density (or maneuverability), and traffic flow. The most significant finding was that it is traffic flow that matters to tractor-trailer driver s, not density. Other important items that affected the perceived qua lity of service included weather, attitudes toward other drivers, and road rage. Some of the freeway conditions discu ssed included the condition of the road surface, lane restrictions, lane width, lane markings, and signage. These factors did not come up in the previous focus groups with freeway automobile commuters. The tractortrailer drivers contende d that trucks have been getting longer and wider at the same time lanes are getting narrower. The condition of the road surface was not the only concern to tractor-trailer drivers, but traffic conditions were also important. A steady traffic flow within an acceptable range was more important to them than actual traffic density. It was clear from the focus group discussion that although tractor trailer drivers are concerned with some of the same characteristi cs of a freeway journey as are automobile

PAGE 25

15 commuters, they also have some different concerns. Urban freeway commuters primary concern was travel time. Steadiness of the tr affic flow was more important in terms of truckers perception of the quality of service. Three other issues were also important to these truckers for their overall opinion on trip quality. The first was simply the phys ical condition of the roadways. The second was maneuverability. Urban commuters seemed content to stay in one lane. Meaning, maneuverability was not important in the urba n context due to higher traffic density. The third trucking issue was the safety of the trip. Safety was also an important concern with urban commuters, although the nature of th eir concerns differed. This paper was informative because it went on to compare and c ontrast tractor-trailer drivers perceptions with factors that are important to freeway automotive commuters. In a study by Hostovsky et al. [11] focus group participants identi fied three themes important for rural freeways LOS. These thr ee themes included the ea se of rural driving (i.e. low density), predicable travel time and steady travel speed, and ample maneuverability on rural freeways. There was also concern about safety related problems on rural freeways, such as problems associated with the isolated nature of rural freeways in association with weather conditions. Secondary rura l freeway themes included aesthetics, speeding, the discomforting presen ce of trucks, and the need for better commuter information. This is an important paper because it showed that rural freeway drivers consider other factor besides density in rating the quality of service of their trip. A paper by Nakamura et al. [12] summarizes a study performed along the Tomei Expressway in Japan on November 27, 1998. In this study, traffic fl ow conditions along a section of a basic intercity motorway were evaluated from the drivers viewpoint. The

PAGE 26

16 study consisted of a field driving survey that wa s intended to collect data on the degree of drivers satisfaction under various uncongested traffic flow conditions along a rural motorway. The major concern of this study was to quantitatively analyze the interrelationship between traffic flow conditions (traffic volume, speed and lane utilization ratio), drivers perceptions (uti lity and degree of sati sfaction) and driving behaviors (lane changing, adjustment of acceler ation and adjustment of spacing) [12]. The objective of the field driving survey wa s to collect data on the perception of drivers and to observe driving behavior under various uncongested traffic flow conditions. Twenty-two subject vehicles drov e themselves from the on-ramp to the offramp and back along the Tomei Expressw ay. A video camera was mounted on the navigators seat of each vehicl e to have the front views recorded. After every trip the subject driver was asked to answer a survey a bout the traffic flow c onditions. A total of 105 surveys were collected. Each traffic flow condition was evaluated using a scale from one to five. The data collected for driving behavior was the number of lane changes, elapsed travel time by lane, and percentage of time spent following another vehicle. It was found that the factor that most strongly affected the degree of drivers satisfaction was the traffic flow rate. The number of lane changing, the percentage of time spent following, and the driving experience al so affected the evaluation of the traffic condition. In addition, setting the LOS ba sed on the average degree of drivers satisfaction was tried and was compared w ith the conventional LOS measures. The results suggest that the curre nt traffic conditions on Japanese motorways hardly satisfy drivers. Meaning, if highways are designed employing the volume to capacity (v/c) ratio that is set only from the drivers degree of satisfaction it woul d take an enormous

PAGE 27

17 monetary investment. This paper was helpful not only because it seeks users opinions but because having participants drive al ong rural freeways was one of the options considered in the alternative approaches eval uation. A few things to consider about this study are that this study was performed in Japan. Also, it collected 105 surveys from only twenty-two subjects. In a study initiated by the Wisconsin Department of Transportation (WDOT) [13] video segments of traffic operations at selected points along a highway were shown to subjects to try and determine what factors we re important to motorists. Survey forms were also used to gather data and correlat e it with field measurements of operational measures of effectiveness (MOE). In addi tion, qualitative input was sought from road users, using focus groups to gain further insights. Survey respondents were shown 50second video clips for six different four-lane rural freeways. Field-measurements were made to determine volume and speed for the video segments shown to the subjects. Density was determined by calculation. The results showed that density and speed were both found to be significantly related to the subjective rating of overa ll quality of flow. This was a helpful paper because it performe d a similar study (by showing video clips to participants and asking their op inion of the roadway conditions) to the type of research that will be performed. Background Survey A study conducted at the University of Fl orida [4] was performed hoping to obtain preliminary information about what factors ar e important to drivers when evaluating the quality of their trip on a rura l freeway. After considerati on of several potential methods for obtaining direct travel er input (i.e., focus groups, post simulation/video review, interview, etc.), it was decided to use a survey -based approach. One of the intents of this

PAGE 28

18 survey-based approach was to help guide a more focused follow-on study that might make use of an alternate method that will onl y allow for a relatively small sample size. Two-hundred and thirty-three travelers were su rveyed. The most important factors to drivers were decided on by the percentage of time they were ranked first, second, or third by drivers. The surveys concluded that the most important factor to drivers was the ability to consistently maintain their desire d travel speed. This was in the top three 64.3% of the time. The second ranked factor was the ability to change lanes and pass other vehicles easily. This was part of the top three responses 33.3% of the time. The third most important factor to drivers was the ability to travel at a speed no less than the posted speed limit. This was in the top three 33.0% of the time. These results show that although density is important to drivers it is not the most important factor, drivers consider other factors besides density in the quality of serv ice of their trip. Conclusions A recent survey of customer-oriented practices of highway agencies pointed out that a customer-oriented perspective not onl y increases the quality of performance but also helps build and sustain necessary public support for transpor tation programs [14]. This literature review has shown while it may not be possible to evaluate how drivers perceive the quality of their trip with extreme precision and c onfidence, it is still possible to arrive at useful conclusions which can he lp the decision makers more closely reflect user perceived quality of service. HCM Freeway Analysis Methodology Overview Urban and Rural Freeway Overview The HCM defines a freeway as a multilane, divided highway with a minimum of two lanes for the exclusive use of traffic in each direction and full control of access

PAGE 29

19 without traffic interruption [1]. The service measure used to determine LOS for freeways is density (passenger cars per mile per lane). Density is a function of both speed and flow rate, which incorporates consideration of freedom to maneuver, driver comfort and convenience. Applicable Thresholds for Urban and Rural Freeways LOS has evolved throughout each revision of the HCM. The original HCM used a basic three-point scale to define level of capacity. In 1963 level of capacity was abandoned for the LOS concept. In 1965 a si x point LOS definition was introduced in the HCM, LOS A LOS F. In 1985 the six point LOS scale was redefined to include traffic density (vehicles per unit length of roadway) as the measure for defining LOS for basic freeway sections. Th is is still used today. Not only do rural and urban freeways have the same service measure, density, but they also have the same limits for LOS. LO S thresholds for a basi c freeway segment are summarized below. LOS Density Range (passenger cars/mile/lane) (pc/mi/ln) A 0-11 B > 11-18 C > 18-26 D > 26-35 E > 35-45 F > 45

PAGE 30

CHAPTER 3 RESEARCH APPROACH Alternative Approaches Evaluation The objective of this study was to develop and test a method by which driver perceptions of LOS for varying roadway conditions can be collected. It is hoped that this method will provide the means to develop a procedure that will lead to specific recommendations for rural freeway LOS criteria and thresholds. An alternative approaches evaluation was created to weigh the pros and cons for methods of collecting drivers input about their driving experiences on rural freeways. Methods of data collection evaluated for this project included: focus groups, in-field surveys, videos from an overhead perspective, videos from a drivers perspective, a driving simulator, driving test participants on rural freeways, and having test participants drive themselves on rural freeways. These are now described in more detail. Focus Group: This method would consist of recruiting individuals to participate in group discussions with a facilitator to guide the discussion in a constructive manner such that the participants reveal the most important factors to them in evaluating quality of service. In-Field Survey: This approach would consist of developing a more focused survey based on the results from the University of Floridas study [4]. The survey would solicit driver opinions about various factors related to the perception of the quality of their trip on a rural freeway. Videos From an Overhead Perspective: This method would consist of having test participants review pre-recorded traffic scenes from an overhead perspective and then answer survey questions related to the observed video scenes. Videos From a Drivers Perspective: This approach would consist of test participants reviewing pre-recorded traffic scenes from a drivers perspective and then answering survey questions related to the observed video scenes. 20

PAGE 31

21 Driving Simulator: This approach would consist of having test participants drive sections of a simulated rural freeway in which traffic and roadway conditions are systematically varied. Specific LOS thresholds would be determined from observed driving behavior and questionnaires administered either during or after the simulation exercises. Driving Test Participants Around on Rural Fr eeways: This approach would consist of having test participants ride in a ve hicle driven by a research official, during which the participants would fill out a tr ip quality questionnaire, with responses being dependent on the given roadway and traffic conditions being experienced at that time. Recruiting Test Participants to Drive on Rural Freeways: This method would consist of having test subjects drive vehicl es in the traffic stream and provide the researchers with real-time feedback either instrumented or verbal. A matrix was developed for these seven diffe rent methods of data collection. This matrix is shown in Table 3-1 The matrix had nine diffe rent criteria on which each approach was judged. For each criterion, ea ch approach was evaluated with a low, medium, high, or not applicable (NA) ra nking, depending on how the criterion related to the approach. The nine criteria chosen for this matrix were: number of test subjects at once, control over experimental conditions, li ability issues, responses while conditions still fresh in mind, supplementa l data collection, drivers pe rspective, concern of motion sickness, ability to relate to actual driving s ituation, and being directly in control of the driving situation. Thes e criteria are now described in more detail. Multiple Test Subjects at Once: This criterion evaluated the number of test subjects that could participate at one time. Control Over Experimental Conditions: This evaluated whether the researchers would be able to control the experimental conditions (i.e. dens ity, percentage of trucks) at the time the subjects participated in the study. Liability Issues: This criterion dealt w ith the liability concerns for the test participants and/or researchers at the time the study was prepared and conducted. Responses While Conditions Still Fresh in Mind: This criterion evaluated if the responses from the subjects would occur during, right after, or a considerable amount of time after a rural freeway trip.

PAGE 32

22 Supplemental Data Collection: This cr iterion addressed if supplemental data collection (i.e. density, percentage of trucks ) would be necessary and if so to what extent. Drivers Perspective: This criterion ev aluated if the approach would give the subject a perspective as if he/she were the driver. Concern of Motion Sickness: This cr iterion addressed th e possibility of participants getting moti on sickness during the study. Ability to Relate to Actual Driving Situation: This criterion evaluated the subjects ability to relate this approach to a real world driving situation. Being Directly in Control of the Driving Situation: This criterion evaluated if the participant was able to cont rol the movements made during the study. It explores if the participants were able to drive ho w they would normally drive on the rural freeway. The in-vehicle video approach was thought to offer the best compromise between a realistic driving experience, study efficiency, and liability c oncerns. As the literature review has shown, video studies in laboratory situations have been successfully used in traffic perception studies for at least 40 years. Video Data Collection Method Since it was concluded to show test participants videos as if they were the driver, a data collection method was developed for this task. The development of this method consisted of six parts: vehicl e instrumentation, determinatio n of data collection sites, collecting video data, composition of video clips, data collection, and a pilot test. Vehicle Instrumentation Again, the objective was to create video clips of rural freeway driving conditions from the drivers perspective. It was d ecided to capture the fo llowing fields-of-view (FOV) from within the vehicle: Front windshield and rear-view mirror Side-view mirror Speedometer

PAGE 33

Table 3-1: Study Approach Alternatives Matrix Number of Test Subjects at Once Control over Experimental Conditions Liability Issues Responses While Conditions Still Fresh in Mind Supplemental Data Collection Drivers Perspective Concern of Motion Sickness Ability to Relate to Actual Driving Situation Being Directly in Control of the Driving Situation Focus Group Med NA Low Low Low Low NA Low NA In-Field Surveys High Low Low High Low High NA Med High Videos from an Overhead Perspective Med High Med High High Low Low Low Low Videos from a Drivers Perspective Med High Med High High Med Med Med Low Driving Simulator Med High Low High Low Med High Med High Driving Test Participants on Rural Freeways Low Low High High High Med Med High Low Having Test Participants Drive on Rural Freeways Low Low High High High High Low High High 23

PAGE 34

24 The vehicle used for instrumentation was a Chevrolet Venture mini-van. The vehicle is shown in Figure 3-1. A total of three video cameras were used. Each video camera was attached to a VCR. A video monitor was attached to each VCR such that a visual confirmation could be made of each cameras FOV. A microphone was also connected to the audio input of one of the VCRs. Portable 12-volt batteries were used as the power source for all the equipment. The front-view camera was set up on a pole between the two front seats. It was attached to the drivers armrest. This is shown in Figure 3-2. The side-view camera was attached to a pole between the drivers seat and door. This is shown in Figure 3-3. The third camera was positioned on the dashboard to capture the speedometer. Duct tape was used to cover the instrument panel to help reduce the glare. This is shown in Figure 3-4. An equipment connection schematic is shown in Figure 3-5. Figure 3-1: Test Vehicle Data Collection Site Selection The rural freeway sites used were all within Florida. It was desired to obtain actual traffic conditions at the time the video data collection runs were made. Thus, one of the requirements in site selection was the presence of inductance loop detectors on the segment. The Florida Department of Transportation (FDOT) maintains an extensive network of inductance loop detectors (ILD) on the Florida Intrastate Highway System (FIHS). There are over 7,500 detector stations on the states highways. There are two types of ILD stations, telemetered and portable. The telemetered stations were preferred

PAGE 35

25 Figure 3-2: Front-View Camera Set-up Figure 3-3: Side-View Camera Picture Figure 3-4: Speedometer Camera Picture Figure 3-5: Equipment Connection Schematic because they record data on a continuous basis (365 days a year) and are permanent stations. The portable stations require a traffic data recorder to be installed in a roadside cabinet (adjacent to the ILD station). Each ILD station has a unique identification number assigned to it. The data collected at the ILD stations is compiled every year and made available on the Florida Traffic Information (FTI) CD [15]. Some other

PAGE 36

26 information that the FTI CD provides is the number of lanes in each direction, the site type (telemetered or portabl e), a site description, the Annual Average Daily Traffic (AADT) information, the percentage of trucks and the peak hour in each direction. All the freeway locations w ith telemetered loop detector s in the state of Florida are shown in Figure 3-6 They are also listed in Appendix B with their site descriptions and AADT information. Not all of these statio ns are located on rural freeway segments, so Geographic Information System (GIS) software was used to help determine suitable locations. Eight specific site locations we re then decided upon. These eight locations can be seen in Figure 3-7 A more detailed breakdown of the traffic conditions at these sites can be found in Table 3-2. Appendix C provides graphical information for these final eight site locations. Both two-lane and three-lane rural freeway sections were selected. South Florida was generally not considered due to limited number of rural locatio ns and long distance from Gainesville. Also, the Panhandle wa s not considered due to driving distance reasons. The Turnpike site was chosen fo r its rolling terrain. The characteristics possessed by these eight site locations addr ess the top four concerns to motorists according to the survey conducted at the Univer sity of Florida [4]. These four items addressed are: ability to maintain travel spee d, ability to change lanes, ability to exceed the posted speed limit, and the percentage of heavy vehicles. It was desired to collect a good range of traffic and roadway conditions. As many different LOS values between A and E as possible was desired. This required the experimenters to try to find low, medium, and heavy traffic conditions for these rural freeways. The FTI CD provided the times for peak hour traffic. These times can be seen

PAGE 37

27 in Table 3-2 In order to determine the LOS of th e rural freeway at the time the roadway was filmed, data that was reco rded by the loop detectors would need to be obtained from the FDOT. The FDOT normally has the loop detectors programmed to report the traffic conditions in one-hour time intervals. Since traffic information for the exact time the loop detector was passed would be pref erred, the FDOT reprogrammed their loop detectors to report the traffic data in five-minute intervals. Video Data Collection Runs After coordination on the repr ogramming of the telemetered data collection stations for five-minute intervals, the actual data collection driving runs were made. Three different drivers made drivi ng runs. All used the same basic driving strategy, which consisted of attempting to drive five mi/hr over the posted speed limit (that is, 75 mi/hr) and generally staying in the right lane unless it was necessary to move to another lane to stay at the desired travel speed (such as a passing maneuver). Each rural freeway was videotaped for a pproximately ten miles before and after the loop detector to ensure consistent traffic conditions. The microphone was used during the experiment to record time, date, and location info rmation on the audio track of one of the videotapes. Each time a loop dete ctor was passed, the route, site number, direction, and time of day was spoken into the microphone. The first day of filming occurred on Monday, November 3, 2003. The sites filmed were along I-75 and I-10 (site numbers 290320, 290269, and 299936). The segment containing the I-75 loop detector (site num ber 290320) was videotaped in the northbound direction at 3:24 p.m., 4:24 p.m ., and 5:33 p.m. The same segment was videotaped in the southbound direction at 4:18 p.m and 5:17 p.m. The segmen t containing the furthest west loop detector along I10 (site number 290269) was vi deotaped in the eastbound

PAGE 38

28 Figure 3-6: Possible Site Locations

PAGE 39

29 Figure 3-7: Final Site Location Map

PAGE 40

Table 3-2: Final Site Location Information Table # Route Site Number # of Lanes Each Direction Truck Lane Restraint Truck AADT Truck % AADT Peak Hour Volume (AADT*K*D) Peak Hour Direction 1: N,E Peak Hour Direction 2: S,W Mile Post 1 I-10 290269 2 No 5,228 25.89 20,192 1,661 6 p.m. 5 p.m. 302-301 2 I-10 299936 2 No 5,502 27.12 20,287 1,446 6 p.m. 5 p.m. 312-311 3 I-75 290320 3 Yes 11,685 25.89 45,131 3,347 4 p.m. 5 p.m. 428-429 4 I-95 730292 2 No 5,072 8.94 56,736 3,213 5 p.m. 4 p.m. 288-289 5 I-75 269904 3 Yes 12,823 21.85 58,687 4,182 1 p.m. & 4 p.m. 4 p.m. 376 6 I-75 189920 2 No 7,016 17.99 39,000 2,356 Unavailable Unavailable 325-324 7 Turnpike 970428 2 No 4,186 12.34 33,926 1,950 6 p.m. 6 p.m. 281 8 I-75 140190 2 No 13,688 19.36 70,702 3,569 6 p.m. 8 a.m. 278-279 30

PAGE 41

31 direction at 3:30 p.m., 4:32 p. m., and 5:40 p.m. This segment was videotaped in the westbound direction at 4:09 p.m. and 5:08 p.m. The last site recorded on November 3 rd was the segment containing the east loop detector on I-10 (site number 299936). This segment was videotaped in the eastbound direction at 3:40 p.m., 4:39 p.m., and 5:48 p.m. This segment was videotaped in the westbound direction at 4:02 p.m., 5:01 p.m., and 6:09 p.m. The second day of filming was to record th e loop detector on I-95 just south of Palm Coast (site number 730292). This segment was videotaped on Tuesday, November 4, 2003. The segment was videotaped in th e northbound direction at 12:55 p.m., 1:21 p.m., 1:46 p.m., 5:04 p.m., and 5:35 p.m. The southbound direc tion of this segment was videotaped at 1:13 p.m., 1:38 p.m., 4:56 p.m., and 5:22 p.m. The third day of filming was to capture th e Tampa and Orlando site locations. This occurred on Wednesday, November 5, 2003. The segments videotaped were loop detectors along I-75 by Micanopy (site number 269904), by Wildwood (site number 189920), by County Road 54 (site number 140190), and along the Turnpike (site number 970428). The first segment, by Micanopy (sit e number 269904), was videotaped heading northbound at 11:05 a.m. and 11:20 a.m. This segment was videotaped in the southbound direction at 11:01 a.m., 11: 16 a.m., and 11:32 a.m. The segment by Wildwood (site number 189920) was videotaped heading nor thbound at 12:19 p.m. and southbound at 12:13 p.m. and 12:29 p.m. The segment south of County Road 54 was videotaped heading northbound at 1:58 p.m. and 2:17 p.m. and southbound at 2:11 p.m. The segment along the Turnpike (site number 970428) was videotaped heading northbound at

PAGE 42

32 4:18 p.m. and 4:50 p.m. The southbound directio n of this segment was videotaped at 3:21 p.m. and 4:33 p.m. Filming this day was cut short due to weather conditions. The fourth, fifth, and sixth days of filming recorded the same sites as the third day did. The fourth day was Friday, November 7, 2003. The segment near Micanopy (site number 269904) was videotaped in the nort hbound direction at 11:55 a.m., 12:13 p.m., and 4:48 p.m. This segment was videotaped in the southbound dire ction at 12:09 p.m. and 12:28 p.m. The segment near Wildwood (site number 189920) was videotaped heading northbound at 1:16 p.m. and 1:34 p.m. and it was videotaped southbound at 1:09 p.m. and 1:28 p.m. The Turnpike segment (site number 970428) was videotaped heading northbound at 3:20 p.m. and southbound at 2:48 p.m. and 3:28 p.m. The fifth day of filming was Sunday, November 9, 2003. Th e segment by County Road 54 (site number 140190) was videotaped in the northbound direct ion at 3:00 p.m. and the segment near Wildwood (site number 189920) was videotaped heading northbound at 3:35 p.m. The sixth day of filming occurred on Friday, November 21, 2003. The segment videotaped on this day was the one near County Road 54 (site number 140190). This segment was videotaped in the northbound direction at 6: 37 a.m., 6:48 a.m., 6:59 a.m., 7:12 a.m., 2:18 p.m., 2:28 p.m., and 2:39 p.m. This segment was videotaped in the southbound direction at 6:41 a.m., 6:52 a.m., 7:04 a.m., 7:15 a.m., 2:11 p.m., 2:22 p.m., and 2:32 p.m. A table of these site locations and dr iving times can be seen in Appendix D Video Clip Composition The survey participants were shown a vi deo that consists of a single composite scene of the three views recorded with the vi deo cameras in the mini-van. The front-view video was the main part of the final video. The front-view video showed the drivers front-view with the rear-view mirror in th e upper middle. The side-view mirror video

PAGE 43

33 was placed in the upper left corner. The speedometer video was shown to the right of the rear-view mirror. The video editing software program, Adobe Premiere [16], was used to create these composite videos. A composite video screenshot is shown in Figure 3-8. Each video was approximately 2 to 2 minutes in length. This length was chosen because the researchers felt that it was a good compromise between consistent traffic conditions experienced on the rural freeway at the time of filming and the survey participants attention span. Figure 3-8: Composite Video Screenshot Loop Data Collection In order to determine the LOS of the traffic conditions when the video runs were made based upon the HCMs density criterion, loop data was simultaneously recorded with the video segments. The loop data collected included speed, volume, and vehicle classification. The telemetered loop detectors were collecting traffic information while the rural freeways were being filmed. The FDOT was able to provide this traffic information to the researchers in five-minute intervals. The information provided by the FDOT included a traffic count file, a classification file, and a speed file for the eight

PAGE 44

34 telemetered site locations used in this st udy. A summary table was created to show the traffic conditions at the times the loop detector s were passed. This table can be seen in Appendix E From this data the density and heavy ve hicle percentage for the site location in the five-minute period in which the segment containing the loop detector was videotaped was calculated. It needs to be mentioned that th e conditions seen in the video scene could still vary substa ntially from the average five-minute conditions calculated from the collected ILD data. Based on these calculations, nine 2 to 2 minute video clips were chosen to be used for the pilot test. The criteria possessed by the nine video clips selected were: 1. Low density, two lane 2. Low density, two lane, rolling terrain 3. Low/Medium density, three lane 4. Medium density, two lane 5. Medium density, two lane, rolling terrain 6. High density, two lane 7. Low heavy vehicle percentage, three lane 8. Medium heavy vehicle percentage, two lane 9. High heavy vehicle percentage, two lane The traffic information, including the LOS ba sed on the HCMs density criteria, for these nine video clips, is shown in Table 3-2 Pilot Test The main purpose of this pilot test was to see if the developed method could be successful at obtaining useful driver perception data on rural freeway LOS. The researchers wanted to determine if the pilot test participants felt they were able to relate to the videos from a drivers perspective. The researchers also wanted to find out if this was a realistic method and what improvements could be made to make this study more accurate.

PAGE 45

35 Pilot test procedure Survey participants for the pilot test were University of Florida civil engineering undergraduate students. Since these participants had just completed the introductory transportation engineering course, they were familiar with the freeway LOS methodology. Because of the video format, mu ltiple test participants could be run simultaneously while allowing the researchers control over the test conditions. As many participants as the room allowed to have a clear view of the screen could be tested at the same time. The video screen was placed at approximately sitting eye level so the participants were looking at th e screen as if they were looki ng out of a cars windshield. Seven students participated in the pilot test The subjects were shown nine 2 to 2 minute videos. Before viewing the videos, te st participants were given only the basic instruction to imagine themselves driving th e vehicle in these conditions, and to make their trip quality ranking based on how they would drive in those conditions, not necessarily how the driver of the instrumented vehicle was driving. The conditions shown by the video clips were representative of an extended rural freeway trip. After each video, the participants were asked to fill out a survey. Survey A test survey was developed for use with th e pilot test. The main objective was to identify the general suitability of the surv ey and any potential improvements. Test participants gave a trip quality ranking after viewing each video. Subjects were not persuaded to use density as the criteria for ra ting the trip quality. They were simply told to just rate the quality of the trip. Trip quality was evaluated us ing six classifications: excellent, very good, good, fair, poor, and very poor. These classifications were generally

PAGE 46

Table 3-2: Loop Detector Traffic Information Clip No. Date (2003) Site Time Dir. Total Volume Avg. Speed 5 min vol veh/hr/ln Density Total 24 Hr Vol 24 Hr Buses 24 Hr Trucks 24 Hr HV 24 Hr % HV LOS 1 11/3 299936 15:40 EB 23 74.1 31.0 186 2.4 8393 15 2504 2519 30.0% A 2 11/21 140190 7:04 SB 122 66.1 301 1806 26.1 42490 88 9113 9201 21.7% D 3 11/21 140190 7:12 NB 71 73.8 164 984 13.8 44492 68 6999 7067 15.9% B 4 11/5 269904 11:16 SB 32 66 115 460 6.5 27737 83 6434 6517 23.5% A 5 11/7 269904 12:13 NB 68 78.1 209 836 11.3 18724 48 5111 5159 27.6% B 6 11/4 730292 12:55 NB 52 78.4 132 792 10.9 no data available A 7 11/4/ 730292 13:21 NB 49 76.8 119 714 9.9 no data available A 8 11/7 970428 14:48 SB 61 71.8 113 678 9 20782 65 4036 4101 19.7% A 9 11/7/ 970428 15:20 NB 89 84.5 164 984 12.5 23811 59 3000 3059 12.8% B 36

PAGE 47

37 intended to correspond to LOS ranking levels A-F. The LOS from the pilot group was then compared to the density-based LOS cal culated in accordance to the HCM. The survey completed by the test participants can be seen in Appendix F

PAGE 48

CHAPTER 4 RESULTS The main purpose of this project was to see if driver perceptions of LOS for varying roadway and traffic conditions could be collected using videos from a drivers perspective. Another desired result was to find out what improvements could be made to make this study more accurate. This chapter consists of four sections. These sections are: pilot test participant feedback on video clips pilot test participant feedback on survey instrument factors that were important in evaluating trip quality a comparison of the HCMs LOS evaluation to the LOS rankings from the pilot test Pilot Test Participant Feedback on Video Clips The last question on the survey given to the participants of the pilot test asked how they would rate this exercise in terms of its ability to give them a feel for the traffic and roadway conditions they would experience if they were actually traveling. The choices available were excellent, very good, good, fair, poor, and very poor. Out of the seven participants, three ranked this method very good, another three ranked this method good, and one participant ranked this method as fair. Some suggestions and recommendations given by the participants were to add sound, improve the video quality, videotape different times of the day and varying weather conditions, and videotape more during peak hour traffic conditions. Two complaints about this method were that it was hard to determine pavement quality and it was hard to compare two-lane rural freeways to three-lane rural freeways. 38

PAGE 49

39 Pilot Test Participant Feedback on Survey Instrument One of the main objectives of the pilot test was to identify the general suitability of the survey and any potential improvements that could be made to improve the survey. The participants of the pilot test did not offer any comments on how to improve the survey. The survey provided the researchers with factors that the participants felt affected their trip quality which was the intent of the survey. Factors That Were Important in Evaluating Quality of Service After the participant evaluated each video clip for the quality of the trip (i.e. excellent, very good, etc.) they were asked to explain what factors led them to give that ranking. Since there were seven participants and nine video clips there were a total of 63 surveys (7 participants x 9 videos) collected during the pilot test. The factor that was mentioned the most in affecting the trip quality was traffic volume. Thirty-one percent of the comments dealt with traffic volume. Pavement quality was mentioned the second most, over twelve percent of the time. The factors that influenced the participants trip quality are listed below in Table 4-1, along with the percentage of time it was mentioned. Table 4-1: Pilot Test Participants Factors Affecting Trip Quality Factors Influencing Trip Quality % of Time Mentioned Traffic Volume 31.0% Pavement Quality 12.6% Ability to Maintain a Constant Speed 9.2% Percentage of Trucks 9.2% Ability to Pass/Change Lanes 8.0% Weather Conditions 7.5% Number of Lanes 4.6% Other Vehicles Speeding By 4.0% Drive at Speed Limit 2.9% Straight Roadway (Easy to See) 2.9% Drive Above Speed Limit 2.9% Cars Following Closely Behind 1.7%

PAGE 50

40 Lane Width 1.1% On/Off Ramps 1.1% Shoulders 1.1% Scenic 0.6% A Comparison of the HCMs LOS Evaluation to the LOS Ratings From the Pilot Test The LOS rankings from the pilot test were calculated by two methods. The first method was the average response. Whichever category had the most responses was the average survey ranking. If the majority of the seven participants rated the video clip excellent it received a LOS rating of A. If the majority of the seven participants rated the video clip good it received a LOS rating of C. The second method was the modal response. This method was calculated by converting the ranking to a numerical scale first, then averaging the rankings, and finally converting the average ranking back to a LOS letter. The participants results were then compared to the LOS calculated by the HCM. The range of the participants responses for trip quality was also noted. Out of the nine video clips, three were rated the same as the HCMs density-based LOS. Five other clips were rated one letter grade lower than the HCMs density-based LOS ranking. The last video clip was rated three letter grades lower than the HCMs density-based LOS ranking. Table 4-2 shows the comparison of the HCMs LOS evaluation to the LOS rankings from the pilot test. There was a correlation between some of the factors the researchers were capturing on the video clips and what some of the participants indicated as factors influencing their trip quality. All of the participants mentioned the number of lanes as a factor influencing their trip quality for the video clips containing the three-lane rural freeway. For the video clips that had a HCM density-based LOS of A participants noted the low traffic volume

PAGE 51

41 as a factor influencing their decision. For the video clip that had a HCM density-based LOS of D the participants noted high traffic volume as a factor influencing their decision. The participants also noted percentage of trucks as a factor influencing their decision on the videotape with the highest percentage of heavy vehicles. However, participants did not note percentage of trucks as a factor influencing their decision when there were one or no trucks in the video clip. Pilot test participants did not consider the rolling terrain on the Turnpike video clips as one of the factors influencing their decision. This could have been because the rolling terrain was hard to pick-up in the video clips. Generally, what was seen in the video clips correlates well to the five-minute loop data collected from the telemetered loop detectors. However, one of the segments, clip number six, appeared to have a more significant amount of traffic at the time it was filmed than the five-minute loop data indicated. In this case it could have been that the five-minute interval was not small enough. Most of the data collected for the five-minute interval could have occurred in the minute the segment was videotaped. As the other eight video clips have shown, they accurately represent the traffic conditions that are occurring on the rural freeway as indicated by the five-minute loop data.

PAGE 52

Table 4-2: Comparison of Results Clip No. Date Site Time Direction Density (pc/hr/ln) 24 Hr % HV HCM LOS Avg. Survey Rankings Modal Response Survey Rankings Range of Survey Rankings 1 11/3/2003 299936 15:40 EB 2.4 30.0% A B A/B A/B 2 11/21/2003 140190 7:04 SB 26.1 21.7% D D/E D/E C/E 3 11/21/2003 140190 7:12 NB 13.8 15.9% B C B/C B/C 4 11/5/2003 269904 11:16 SB 6.5 23.5% A A A/B A/C 5 11/7/2003 269904 12:13 NB 11.3 27.6% B B B/C A/D 6 11/4/2003 730292 12:55 NB 10.9 No data A D/E C/D C/D 7 11/4/2003 730292 13:21 NB 9.9 No data A A A/B A/B 8 11/7/2003 970428 14:48 SB 9 19.7% A B B A/C 9 11/7/2003 970428 15:20 NB 12.5 12.8% B C B/C A/C 42

PAGE 53

CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS The purpose of the pilot test was to see if a method had been developed by which driver perceptions of LOS for varying roadway conditions could be collected. The development of this method consisted of accomplishing six parts: vehicle instrumentation, determination of data collection sites, collecting video data, composition of video clips, data collection, and a pilot test. The researchers wanted to identify if the participants were able to relate to the videos from a drivers perspective. The researchers also wanted to find out if this was a realistic method and what improvements could be made to make this study more accurate. Conclusions Vehicle Instrumentation The instrumentation of the vehicle produced the video clips the researchers were hoping for. All three views were captured simultaneously with good quality. The video clips could have been more consistent if the video cameras were permanently mounted in the test vehicle. Having to set up the test vehicle everyday not only took time but it led to inconsistencies in the video cameras FOV. Determination of Data Collection Sites The sites chosen worked well for the purpose of this study. These sites were chosen based on their AADT and percentage of heavy vehicle data from the FTI CD. The researchers were able to capture both two and three-lane rural freeways, low density and medium density conditions, both low and high percentages of heavy vehicles, and 43

PAGE 54

44 flat and rolling terrain. The researchers we re unable to capture the higher range of density conditions. Collecting Video Data For this study, more than one site location was videotaped in a day. This led to the inability to capture all the vary ing traffic conditions at one site. Some sites were filmed in the middle of the day when there wasnt much traffic while other sites were filmed only during peak hour traffic conditions. The be st approach would be to devote at least one day for each site that is going to be vide otaped and capture the full range of density conditions for that particular site. Composition of Video Clips The video editing software program, Adobe Premier, was used to composite the video clips. This software was able to merg e three videotapes togeth er in the layout that the researchers desired. There were two problems with the com position of the video clips. The main problem was when the videot apes were transformed into digital format the digital output would sometimes freeze. This led to the second problem which was synchronization between the three camera view s. If one of the views froze it would become off sync from the other two views. Data Collection The traffic information was provided from the telemetered loop detectors in fiveminute intervals. Generally, what was seen in the video clips correlated well to the fiveminute loop data collected from the telemetere d loop detectors. The alternative to using the telemetered loop data is to determine if there is enough correlation with what can be seen through the video to use what can be seen as the traffic data.

PAGE 55

45 Pilot Test From the pilot test most of the participants were able to relate to the videos as if they were the driver. There were a few s ituations where participants commented that they would have driven differen tly, but for the most part the participants said they were able to relate to being the driv er. Most of the participants also felt that this was a realistic method for what the researchers were trying to achieve. The results of the pilot test showed that subjects LOS rankings were not far from the density-based HCMs LOS ranking. Out of the seven participants, no one used density as the only criterion when rating LOS. All participants considered more than just density. Recommendations The method of showing subjects videos as if they were the driver was successful. The overall video quality used in the pilot te st could be improved. Also, another way to better show pavement quality on the videos should be investigated. Since density was not the only consideration by th e participants in this study, it is recommended that a more detailed study using this resear ch approach be conducted. It is hoped that a more detailed study will provide the means to develop a procedure that will lead to specific recommendations for rural freeway LOS criteria and thresholds.

PAGE 56

APPENDIX A COMMON ABBREVIATIONS HCM Highway Capacity Manual LOS Level of Service TRB Transportation Research Board HCQS Highway Capacity and Quality of Service QOS Quality of Service LF Load Factor v/c Volume to Capacity Ratio AID Average Individual Delay WDOT Wisconsin Department of Transportation MOE Measures of Effectiveness pc/mi/ln Passenger Cars per Mile per Lane NA Not Applicable FOV Field of View FDOT Florida Department of Transportation ILD Inductance Loop Detectors FIHS Florida Intrastate Highway System AADT Average Annual Daily Traffic GIS Geographic Information System FTI Florida Traffic Information 46

PAGE 57

APPENDIX B POSSIBLE SITE LOCATION TRAFFIC INFORMATION

PAGE 58

Site Site Type Description Direction 1 Direction 2 AADT TwoWay "K" Factor "D" Factor "T" Factor 30191 T SR-93/I-75,0.5 MI N OF CR-896,COLLIER CO. 27,043 N 27,573 S 54,616 C 9.88 A 53.63 A 13.05 D 30351 T SR-93/I-75,W OF EVERGLADES BLVD,COLLIER CO. 9,169 E 9,182 W 18,351 C 12.94 A 56.13 A 11.46 A 100106 T SR-400/I-4, UNDER BETHLEHAM RD OVERPASS, HILL CO 48,583 E 50,946 W 99,529 C 8.30 A 54.20 A 19.36 D 100110 T I-275, 1.3 MI E OF HOWARD FRANKLIN BR 62,329 E 62,426 W 124,755 C 8.46 A 55.20 A 8.71 D 100123 T 1-275 AT S. END OF FLORIBRASKA AVE OVERP-BAD LP 65,831 C 9.20 D 53.92 D 8.71 D 100194 T I-75, 0.6 M S OF US301, 2.3 M N OF I-4, HILLS--BAD LOOP 49,231 N 48,660 S 97,891 C 9.10 D 53.50 D 9.35 A 100224 T I-75, 0.9 MI S OF SR60 AT SR618 O.P., HILLSBOROUGH CO 51,358 N 52,487 S 103,845 C 9.20 D 53.92 D 9.82 A 109922 T SR-93/I-275, 0.25 MI N OF FLETCHER AV, TAMPA, HILL CO 66,000 S 9.20 D 53.92 D 8.71 D 109926 T 1-75, 1.25 MI N OF SR-60 (ADAMO DR.), TAMPA, WIM#26 57,374 N 54,369S 111,743 C 9.20 D 53.92 D 7.05 A 120184 T SR-93/I-75, 225' S OF DANIELS PKWY UNDERPASS, LEE 24,674 N 21,993 S 46,667 C 9.73 A 57.77 A 13.05 D 140190 T I-75, 0.6 MI. SOUTH OF SR-54, PASCO CO. UC10/28/94 35,427 N 35,275S 70,702 C 8.99 A 56.15 A 19.36 D 150183 T I-275, 300 YDS S OF THE SB TOLL, PINELLAS-UC11/93 22,171 N 22,848S 45,019 C 9.57 A 52.30 A 5.87 A 170225 T I-75, @ PROCTOR RD OP, 0.7 MI N SR 72, SARASOTA CO 42,732 N 42,863 S 85,595 C 9.83 A 53.85 A 18.10 D 189920 T I-75, 3.5 MI S OF FLORIDA TURNPIKE, WIM#20 39,000 F 10.97 D 55.06 D 17.99 D 269904 T I-75/SR-93, 3 MILES NORTH OF MARION COUNTY LINE 29,922 N 28,765S 58,687 C 12.50 D 57.01 D 21.85 A 290269 T I-10, 0.45 MI EAST OF US41, LAKE CITY OF LAKE CITY BETWEEN I-10 AND 10,164 E 22,340 10,028 W 20,192 C 45,131 12.27 A 67.04 A 25.89 A 290320 T I-75 NORTH US-90 N 22,791 S C 13.62 A 54.44 A 25.89 A 48

PAGE 59

Site Site Type Description Direction 1 Direction 2 AADT TwoWay "K" Factor "D" Factor "T" Factor 299936 T I-10, 50 FT WEST OF CR-250 OVERPASS, LAKE CITY 10,104 E 10,183 W 20,287 C 12.50 D 57.01 D 27.12 A 320112 T I-75, S OF STATE LINE NORTH OF S.R. 143 18,515 N 18.683S 37,198 C 15.17 A 54.98 A 29.93 A 360317 T I-75, SB SHOULDER, 0.35 MILES N OF WILLIAMS RD. 37,579 N 26,937 S 74,516 C 11.69 A 54.81 A 17.99 D 370238 T I-10, 0.15 MILES WEST OF CR 136 12,725 E 12,621 W 25,346 C 11.89 A 53.84 A 22.64 A 370352 T I-10, SUWANNEE CO, 1.5 MI W OF ELLAVILLE SCALES 8,124 C 12.50 D 57.01 D 29.61 A 480156 T I-10, 1.5 MI WEST OF US-90 16,073 E 14,564 W 30,637 C 10.56 A 58.54 A 26.51 D 480560 T SR-8/I-10, 1 MILE OF SR-291/DAVIS HWY (RTMS) 0 E 20,538 W 0 C 9.64 D 51.15 D 14.60 D 489924 T I-110, 1 MI S OF I-10, PENSACOLA, WIM#24--UC 10/94 26,390 N 26,051S 52,441 C 9.64 D 51.15 D 4.05 A 500220 T I-10, 250 FT W OF CR-268 OVERPASS, GADSDEN CO. 13,664 E 13,337 W 27,001 C 11.68 A 54.72 A 18.22 A 530218 T I-10, 1 MI. EAST OF US-231, JACKSON CO. -UC 11/94 10,821 E 10,768 W 21,589 C 12.90 A 52.53 A 26.51 D 549901 T SR-8/I-10,0.66 MI E OF CR-257,JEFFERSON CO. 12,478 E 12,527 W 25,005 C 13.69 A 52.46 A 22.78 A 550208 T MISSION RD,NORTH OF I-10,TALLAHASSEE,LEON CO. 4, 896 N 4,916 S 9,812 C 11.18 D 57.23 D 5.3 D 550304 T SR-8/I-10,1 MI W OF THOMASVILLE RD U/P,LEON CO. 28,625 E 28,275 W 56,900 C 9.64 A 51.15 A 12.44 A 570318 T SR-8/I-10,@ANTIOCH RD O/P,OKALOOSA CO. 10,347 E 10,504 W 20,851 C 10.65 A 52.21 A 19.68 A 609928 T SR-8/I-10,1.3 MI WEST OF BOY SCOUT RD,WALTON CO. 9,590 E 9,526 W 19,116 C 12.05 D 55.16 D 21.33 A 610152 T SR-8/I-10,AT CR-273,SE OF CHIPLEY,WASHINGTON CO. 9,320 E 9,355 W 18,675 C 35,850 12.41 A 58.45 A 21.51 A 700134 T SR-9/I-95,3.34 MI. S. OF SR-514,BREVARD CO. 17,780 N 18,070 S C 10.97 D 55.06 D 14.88 A 49

PAGE 60

Site Site Type Description Direction 1 Direction 2 AADT TwoWay "K" Factor "D" Factor "T" Factor 700322 T SR-9/I-95,0.9 MI S OF AURANTIA RD U/P,BREVARD CO. 14,566 N 14,601 S 29,167 C 11.32 A 54.06 A 18.7 A 709919 T SR-9/I-95,3.45 MI S OF SR-514,MALABAR,BREVARD CO. 17, 994 N 17,726 S 35,720 C 10.97 D 55.06 D 16.84 A 729923 T SR-9/I-95,0.75 MI S OF DUNN AVE,JAX,DUVAL CO. 69,500 S 9.06 D 55.55 D 10.57 D 720354 T SR-8/I-10,1 MI E OF MCDUFF AV,JAX,DUVAL CO. 68,110 E 69,028 W 137,138 C 8.83 A 61.55 A 4.15 A 720109 T SR-8/I-10,@CR-217 OVERPASS,E. OF BALDWIN,DUVAL CO. 22,665 E 22,429 W 45,094 C 9.54 A 54.73 A 22.13 A 720171 T SR-9/I-95,0.7 MI N OF UNIVERSITY BLVD,JAX,DUVAL CO 55,386 N 55,381 S 110,767 C 9.11 A 52.57 A 5.77 A 729914 T SR-9A/I-295,3 MI N OF I-10,JACKSONVILLE,DUVAL CO. 31,128 N 30,663 S 61,791 C 9.24 A 52.52 A 14.66 A 729905 T SR-9/I-95,2 MI S OF I-295 S INTERCHANGE,DUVAL CO. 34,405 N 33,453 S 67,858 C 9.06 D 55.55 D 10.57 D 730292 T SR-9/I-95,1.4 MI S OF PALM COAST PKWY,FLAGLER CO. 28,115 N 28,621 S 56,736 C 10.14 A 55.85 A 8.94 D 740132 T SR-9/I-95,2.0 MI S OF GA. STATE LINE,NASSAU CO. 26,031 N 25,761 S 51,792 C 12.5 D 57.01 D 26.82 D 770343 T SR-400/I-4,1.6 MI E OF SR-434,SEMINOLE CO. 60,981 E 60,669 W 121,650 C 8.06 A 52.46 A 6.41 A 790133 T I-95,2.7 MI N OF SR44,@CR44 O/P,VOLUSIA CO. 20,046 N 19,745 S 39,791 C 9.89 A 56.31 A 13.35 A 799906 T ON I-4,169' E OF ENTERPRISE RD O/P,VOLUSIA CO. 0 E 0 W 76,000 F 8.61 D 54.73 D 8.94 D 860163 T SR-9/I-95,@NE 48TH ST,POMPANO BCH,BROWARD CO. 99,069 N 99,696 S 198,765 C 7.83 A 51.22 A 6.83 A 860186 T SR-862/I-595,0.2 MI E OF UNIVERSITY DR,BROWARD CO. 91, 065 E 86,047 W 177,112 C 8.32 A 55.82 A 4.7 A 860357 T SR-93/I-75,2 MI W OF US-27,.6 MI W TOLL,BROWARD CO 10,043 E 48,348 10,026 W 47,387 20,069 C 95,735 12.22 A 58.41 A 13.42 A 870108 T SR-112/I-195,1600' E OF SR-5/US-1,DADE CO. E W C 8.47 A 52.54 A 2.35 D 50

PAGE 61

Site Site Type Description Direction 1 Direction 2 AADT TwoWay "K" Factor "D" Factor "T" Factor 920303 T ON I-4,0.5 MI SW OF ORANGE CO LINE,OSCEOLA CO. 54,281 E 55,142 W 109,423 C 7.63 A 55.88 A 7.42 A 930198 T SR-9/I-95,@SW23RD AVE O/P,1.5 MI S SR-804,PALM BCH 159,000 S 8.56 D 55.24 D 10.8 D 930217 T SR-9/I-95,0.8 MI N OF DONALD ROSS RD,PALM BEACH CO 41,368 N 42,333 S 83,701 C 8.81 A 61.12 A 10.17 A 940260 T SR-9/I-95,0.6 MI S OF SR-68/ORANGE AV,ST LUCIE CO. 55.24 D 42,500 F 8.56 D 10.8 D 940334 T SR-9/I-95,@MARTIN CO LINE,ST. LUCIE CO. 21,475 N 22,348 S 43,823 C 9.28 A 52.79 A 19.94 A Site Type: T = Telemetered; P = Portable AADT Flags: C = Computed; E = Manual Estimate; F = First Year Est; S = Second Year Est; T = Third Year Est; X = Unknown "K/D" Flags: A = Actual; F = Volume Fctr Catg; D = Dist/Functional Class; S = State-wide Default; W = One-Way Road ss; S = State-wide Default; X = Cross"T" Flags: A = Actual; F = Axle Fctr Catg; D = Dist/Functional ClaReference 51

PAGE 62

APPENDIX C FINAL SITE LOCATION DETAILS

PAGE 63

53

PAGE 64

54

PAGE 65

55

PAGE 66

56

PAGE 67

57

PAGE 68

58

PAGE 69

APPENDIX D SITE LOCATIONS AND DRIVING TIMES

PAGE 70

Site # of Lanes Peak Hour Peak Hour # Route Number Direction N, E Direction S, W Direction MP Date Time Direction 1 I-10 290269 2 6 p.m. 5 p.m. 302-301 11/3/2003 3:30 p.m. EB 4:09 p.m. WB 4:32 p.m. EB 5:08 p.m. WB 5:40 p.m. EB 2 I-10 299936 2 6 p.m. 5 p.m. 312-311 11/3/2003 3:40 p.m. EB 4:02 p.m. WB 4:39 p.m. EB 5:01 p.m. WB 5:48 p.m. EB 6:09 p.m. WB 3 I-75 290320 3 4 p.m. 5 p.m. 428-429 11/3/2003 3:24 p.m. NB 4:18 p.m. SB 4:24 p.m. NB 5:17 p.m. SB 5:33 p.m. NB 4 I-95 730292 2 5 p.m. 4 p.m. 288-289 11/4/2003 12:55 p.m. NB 1:13 p.m. SB 1:21 p.m. NB 1:38 p.m. SB 1:46 p.m. NB 4:56 p.m. SB 5:04 p.m. NB 5:22 p.m. SB 5:35 p.m. NB 5 I-75 269904 3 1 p.m. & 4 p.m. 4 p.m. 376 11/5/2003 11:01 a.m. SB 11:16 a.m. SB 11:20 a.m. NB 60

PAGE 71

Site # of Lanes Peak Hour Peak Hour # Route Number Direction N, E Direction S, W Direction MP Date Time Direction 11:05 a.m. NB 11:32 a.m. SB 11/7/2003 11:55 a.m. NB 12:09 p.m. SB 12:13 p.m. NB 12:28 p.m. SB 4:48 p.m. NB 6 I-75 189920 2 325-324 11/5/2003 12:13 p.m. SB 12:19 p.m. NB 12:29 p.m. SB 11/7/2003 1:09 p.m. SB 1:16 p.m. NB 1:28 p.m. SB 1:34 p.m. NB 11/9/2003 3:35 p.m. NB 7 Turnpike 970428 2 6 p.m. 6 p.m. 281 11/5/2003 3:21 p.m. SB 4:18 p.m. NB 4:33 p.m. SB 4:50 p.m. NB 11/7/2003 2:48 p.m. SB 3:20 p.m. NB 3:28 p.m. SB 8 I-75 140190 2 6 p.m. 8 a.m. 276-275 11/5/2003 1:58 p.m. NB 2:11 p.m. SB 2:17 p.m. NB 11/9/2003 3:00 p.m. NB 11/21/2003 6:37 a.m. NB 6:41 a.m. SB 61

PAGE 72

Site # of Lanes Peak Hour Peak Hour # Route Number Direction N, E Direction S, W Direction MP Date Time Direction 6:48 a.m. NB 6:52 a.m. SB 6:59 a.m. NB 7:04 a.m. SB 7:12 a.m. NB 7:15 a.m. SB 2:11 p.m. SB 2:18 p.m. NB 2:22 p.m. SB 2:28 p.m. NB 2:32 p.m. SB 2:39 p.m. NB 62

PAGE 73

APPENDIX E LOOP DATA SUMMARY TABLES

PAGE 74

D ate Tag Site Time Direction Total Volume Avg. Speed 5 min vol veh/hr/ln density 11/3/2003 SPD 290320 (I-75, near I-10) NO DATA AVAILABLE 11/3/2003 SPD 290269 (I-10, near I-75 NO DATA AVAILABLE 11/3/2003 SPD 299936 (I-10, near I-75) 15:40 Eastbound 23 74.1 31.0 186 2.4 16:39 27 74.3 31 186 2.5 17:48 24 71 32 192 2.6 16:02 Westbound 15 76.3 55 330 4.4 17:01 11 78 44 264 3.5 18:09 4 75.5 25 150 2.1 11/4/2003 SPD 730292 (I-95, near Daytona) 12:55 Northbound 52 78.4 132 792 10.9 13:21 49 76.8 119 714 9.9 13:46 39 78.3 120 720 9.8 64 17:04 36 78.2 106 636 8.5 17:35 56 78 141 846 11.6 13:13 Southbound 77 71.5 116 696 9.6 13:38 73 70.6 135 810 11 16:56 75 72.7 147 882 11.9 17:22 91 71.9 147 882 11.7 11/5/2003 SPD 269904 (I-75, near Micanopy) 11:05 Northbound 19 80.9 116 464 6.3 11:20 30 80 136 544 7.3 11:01 Southbound 41 66 144 576 8.4 11:16 32 66 115 460 6.5 11:32 44 65.6 148 592 8.4

PAGE 75

Date Tag Site Time Direction Total Volume Avg. Speed 5 min vol veh/hr/ln density 11/5/2003 SPD 189920 (I-75, s. of Turnpike merge) 12:19 Northbound NO DATA AVAILABLE 12:13 Southbound NO DATA AVAILABLE 12:29 NO DATA AVAILABLE 11/5/2003 SPD 140190 (I-75, near Tampa) 13:58 Northbound 75 73.37 137 822 11.6 14:17 79 74.01 164 984 14 14:11 Southbound 92 67.08 160 960 13.5 11/5/2003 SPD 970428 (Turnpike, near Orlando) 16:18 Northbound 54 78.4 107 642 8.8 16:50 54 79.6 105 630 8.4 15:21 Southbound 45 69.8 73 438 6 16:33 55 69.5 112 672 9.3 11/7/2003 SPD 269904 (I-75, near Micanopy) 11:55 Northbound 48 79.2 180 720 9.6 12:13 68 78.1 209 836 11.3 65 16:48 120 79.1 293 1172 15.5 12:09 Southbound 48 69 195 780 10.8 12:48 39 67.1 163 652 8.9 11/7/2003 SPD 189920 (I-75, s. of Turnpike merge) 13:16 Northbound DATA AT ONE HOUR INTERVAL 13:34 DATA AT ONE HOUR INTERVAL 13:09 Southbound DATA AT ONE HOUR INTERVAL 13:28 DATA AT ONE HOUR INTERVAL 11/7/2003 SPD 970428 (Turnpike, near Orlando) 15:20 Northbound 89 84.5 164 984 12.5 14:48 Southbound 61 71.8 113 678 9

PAGE 76

Date Tag Site Time Direction Total Volume Avg. Speed 5 min vol veh/hr/ln density 15:28 60 72.1 123 738 9.8 11/9/2003 SPD 140190 (I-75, near Tampa) 15:00 Northbound NO DATA AVAILABLE 11/9/2003 SPD 189920 (I-75, s. of Turnpike merge) 15:35 Northbound DATA AT ONE HOUR INTERVAL 11/21/2003 SPD 140190 (I-75, near Tampa) 6:37 Northbound 54 75.5 135 810 11.2 6:48 67 76.4 166 996 13.8 6:59 56 74.4 145 870 12.2 7:12 71 73.8 164 984 13.8 14:18 102 73.5 236 1416 20.2 14:28 158 71.2 274 1644 24.1 14:39 101 74 228 1368 19.2 6:41 Southbound 123 65.3 306 1836 26.7 6:52 124 65.1 297 1782 26 66 7:04 122 66.1 301 1806 26.1 7:15 135 65.5 302 1812 26.3 14:11 88 72 211 1266 17.1 14:22 96 70.8 214 1284 17.4 14:32 89 67.1 227 1362 19.2

PAGE 77

Date Tag Site Time Direction 5 min vol HV veh/hr/ln HV Total 24 Hr Vol 24 Hr Buses 24 Hr Trucks 24 Hr HV 24 Hr %HV 11/3/2003 CLS 290320 (I-75, near I-10) 11/3/2003 CLS 290269 (I-10, near I-75) no data available 11/3/2003 CLS 299936 (I-10, near I-75) 15:40 Eastbound 4 24 Total 17277 31 5057 5088 29.4% 16:39 6 36 17:48 12 72 WB 8884 16 2553 2559 28.8% 16:02 Westbound 21 126 17:01 12 72 EB 8393 15 2504 2519 30.0% 18:09 7 42 11/4/2003 CLS 730292 (I-95, near Daytona) 12:55 Northbound 13:21 13:46 17:04 17:35 13:13 Southbound 13:38 16:56 17:22 no data available 11/5/2003 CLS 269904 (I-75, near Micanopy) 11:05 Northbound 25 150 Total 50443 104 13439 13543 26.8% 67 11:20 34 204 11:01 Southbound 32 192 NB 19320 40 6699 6739 34.9% 11:16 22 132 11:32 34 204 SB 31123 64 6740 6804 21.9%

PAGE 78

Date Tag Site Time Direction 5 min vol HV veh/hr/ln HV Total 24 Hr Vol 24 Hr Buses 24 Hr Trucks 24 Hr HV 24 Hr %HV 11/5/2003 CLS 189920 (I-75, s. of Turnpike merge) 12:19 Northbound Total 35682 78 10504 10582 29.7% 12:13 Southbound NB 16958 30 5393 5423 32.0% 12:29 data at one hour interval SB 18724 48 5111 5159 27.6% 11/5/2003 CLS 140190 (I-75, near Tampa) 13:58 Northbound 29 174 Total 72280 78 12171 12249 16.9% 14:17 23 138 14:11 Southbound 30 180 NB 34775 36 6139 6175 17.8% 68 SB 37505 42 6032 6074 16.2% 11/5/2003 CLS 970428 (Turnpike, near Orlando) 16:18 Northbound 11 66 Total 29694 81 7823 7904 26.6% 16:50 30 180 15:21 Southbound 16 96 NB 43 2748 2791 16:33 16 96 11/7/2003 CLS 269904 (I-75, near Micanopy) 11:55 Northbound 31 186 Total 76912 204 12112 12316 16.0% 12:13 32 192 16:48 30 180 NB 27737 83 6434 6517 23.5% 12:09 Southbound 21 126 12:48 25 150 SB 49175 121 5678 5799 11.8%

PAGE 79

Date Tag Site Time Direction 5 min vol HV veh/hr/ln HV Total 24 Hr Vol 24 Hr Buses 24 Hr Trucks 24 Hr HV 24 Hr %HV 11/7/2003 CLS 189920 (I-75, s. of Turnpike merge) 13:16 Northbound Total 51946 146 9862 10008 19.3% 13:34 13:09 Southbound NB 27449 69 5315 5384 19.6% 13:28 data at one hour interval SB 24497 77 4547 4624 18.9% 11/7/2003 CLS 970428 (Turnpike, near Orlando) 15:20 Northbound 14 84 Total 44593 124 7036 7160 16.1% 14:48 Southbound 11 66 15:28 19 114 NB 23811 59 3000 3059 12.8% 69 SB 20782 65 4036 4101 19.7% 11/9/2003 CLS 140190 (I-75, near Tampa) 15:00 Northbound no data available 11/9/2003 CLS 189920 (I-75, s. of Turnpike merge) 15:35 Northbound Total 45990 128 4484 4612 10.0% NB 19108 69 1668 1737 9.1% data at one hour interval SB 26882 59 2816 2875 10.7%

PAGE 80

Date Tag Site Time Direction 5 min vol HV veh/hr/ln HV Total 24 Hr Vol 24 Hr Buses 24 Hr Trucks 24 Hr HV 24 Hr %HV 11/21/2003 CLS 140190 (I-75, near Tampa) 6:37 Northbound 23 138 Total 86982 156 16112 16268 18.7% 6:48 33 198 6:59 36 216 NB 44492 68 6999 7067 15.9% 7:12 30 180 14:18 31 186 SB 42490 88 9113 9201 21.7% 14:28 42 252 14:39 35 210 6:41 Southbound 98 588 6:52 109 654 7:04 97 582 7:15 104 624 70 14:11 24 144 14:22 30 180 14:32 28 168

PAGE 81

APPENDIX F PILOT TEST SURVEY

PAGE 82

Transportation Research Center About Yourself Gender: Male Female Age: 16 to 25 years 26 to 45 years 46 to 65 years Over 65 years Highest level of education: Some or no high school High school diploma or equivalent Technical college degree (A.A.) College degree Post-graduate degree Approximate annual household income: No income Under $25,000 $25,000 49,999 $50,000 74,999 $75,000 99,999 $100,000 149,999 $150,000 or more Number of years possessing a drivers license: _________ About Your Rural F ree w ay Drivin g Typical number of rural freeway trips made during a month? 1 to 2 3 to 4 5 to 6 7 to 8 9 to 10 11 to 12 Over 12 Typical length of trip made on rural freeway (in miles)? less than 16 miles 16 to 30 31 to 45 46 to 60 61 to 75 76 to 100 101 to 125 126 to 150 151 to 175 176 to 200 Over 200 Vehicle type most often used for rural freeway trips: Sedan Sports car Pick-up truck SUV Mini-van Full-size van RV/Motorhome motorcycle Other _______________ 72

PAGE 83

73 Your Opini ons Rank the overall quality of the trip for the conditions observed in the video clip. Place an X in the appropriate column of the table below. Video Clip Excellent Very Good Good Fair Poor Very Poor List all the factors/reasons that influenced your ranking of the trip quality for this video clip. After listing the factors, please number them from most important to least important _______________________________________________________________________ _______________________________________________________________________ _______________________________________________________________________ _______________________________________________________________________ _______________________________________________________________________ _______________________________________________________________________ _______________________________________________________________________ _______________________________________________________________________

PAGE 84

74 In general, how would the purpose of your trip (such as business, recreational, social) affect the trip quality rankings assigned above? _______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ How would you rate this exercise in terms of its ability to give you a feel for the traffic and roadway conditions you would experience if you were actually driving? Excellent Very Good Good Fair Poor Very Poor

PAGE 85

LIST OF REFERENCES 1. Transportation Research Board. (2000). Special Report 209: Highway Capacity Manual. Transportation Research Board, Washington, D.C., 4 th Ed. 2. Harwood, D., Flannery, A., McLeod, D., Vandehey, M. (July 2001). The Case for Retaining the Level of Service Concept in the Highway Capacity Manual. Presented at the 2001 Transportation Research Board Committee A3A10 Highway Capacity and Quality of Service Midyear Meeting, Truckee, California. 3. Transportation Research Board. (1985). Special Report 209: Highway Capacity Manual. Transportation Research Board, Washington, D.C. 4. Washburn, S. (2003). Facility Performance Model Enhancements for Multimodal Systems Planning. Transportation Research Center at the University of Florida, Gainesville, Florida. 5. Pcheux, K., Tarko, A., Rabbani, E., Hall F. (July 2001). Scouting Party on User Perceptions and LOS Performance Measures. Presented at the 2001 Transportation Research Board Committee A3A10 Highway Capacity and Quality of Service Midyear Meeting, Truckee, California. 6. Sutaria, T.C., and Haynes, J.J. (1977). Level of Service at Signalized Intersections. Transportation Research Record 644, Transportation Research Board, Washington, D.C., 107-113. 7. Highway Research Board. (1965). Special Report 87: Highway Capacity Manual. National Research Council, Washington, D.C. 8. Pcheux, K., Pietrucha, M., Jovanis, P. (2000). User Perception of Level of Service at Signalized Intersections: Methodological Issues. Transportation Research Circular E-C018: Proceedings of the Fourth International Symposium on Highway Capacity, National Research Council, Washington, D.C., 322-335. 9. Hall, F., Wakefield, S., Al-Kaisy, A. (2000). Freeway Quality of Service: What Really Matters to Drivers and Passengers? Transportation Research Record: Journal of the Transportation Research Board, No. 1776, Transportation Research Board, National Research Council, Washington, D.C., 17-23. 10. Hostovsky, C., Hall, F. (2003). Freeway Quality of Service: Perceptions from Tractor-Trailer Drivers. Presented at the Transportation Research Board 82 rd Annual Meeting, Washington, D.C. 75

PAGE 86

76 11. Hostovsky, C., Wakefield, S., Hall, F. (2003). Mitigating Traffic Congestion Impacts: Users Perceptions of the Quality of Transportation Service. Submitted to the Canadian Geographer for publication. 12. Nakamura, H., Suzuki, K., Ryu, S. (2000). Analysis of the Interrelationship Among Traffic Flow Conditions, Driving Behavior, and Degree of Drivers Satisfaction on Rural Motorways. Transportation Research Circular E-C018: Proceedings of the Fourth International Symposium on Highway Capacity, National Research Council, Washington, D.C., 42-52. 13. Pfefer, R. (2000). Toward Reflecting Public Perception of Quality of Service in Planning, Designing and Operating Highway Facilities. Transportation Research Record: Journal of the Transportation Research Board, No. 1685, Transportation Research Board, National Research Council, Washington, D.C., 53-62. 14. Stein-Hudson, K. (1995). Customer-Based Quality in Transportation. NCHRP Report 376, Transportation Research Board, National Research Council, Washington, D.C. 15. Florida Traffic Information 2002. (2002). Florida Department of Transportation, Tallahassee, FL, CD-ROM. 16. Users Guide for Adobe Premiere Pro Software. (n.d.). Retrieved November 17, 2003, from http:// www.adobe.com/products/premiere 17. Transportation Research Board. (1998). Special Report 209: Highway Capacity Manual. Transportation Research Board, Washington, D.C., 3 rd Ed. 18. Kittelson, W. (2001). Highway Capacity and Quality of Service. Presented at the 2001 Transportation Research Board Committee A3A10 Highway Capacity and Quality of Service Midyear Meeting, Truckee, California.

PAGE 87

BIOGRAPHICAL SKETCH I am a 23 year old graduate student at the University of Florida. I received my undergraduate degree of Bachelor of Science in Civil Engineering from the University of Florida in May of 2002. I am currently working on my Master of Science from the College of Engineering, specializing in transportation. 77


Permanent Link: http://ufdc.ufl.edu/UFE0003401/00001

Material Information

Title: An Exploratory data collection approach for the assessment of level of service from a traveler's perspective
Physical Description: Mixed Material
Language: English
Creator: Seager, Kimberly ( Dissertant )
Washburn, Scott S. ( Thesis advisor )
Courage, Dr. ( Reviewer )
Li, Tao ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2004
Copyright Date: 2004

Subjects

Subjects / Keywords: Civil and Coastal Engineering thesis, M.S
Dissertations, Academic -- UF -- Civil and Coastal Engineering

Notes

Abstract: Although the concept of level of service (LOS) for freeways is usually defined in terms of users' perceptions, there have been very few studies that have sought drivers, or passengers' views about what is important to them. The objective of this study was to develop and test a method by which driver perceptions of LOS for varying roadway conditions can be collected. The approach decided upon was to create video clips of rural freeway driving conditions from the driver's perspective. These video clips consisted of the front-view, the rear-view, the side-view, and the speedometer. Each video clip was between 2 to 2 1/2 minutes. In total, there were nine video clips. These video clips were then shown to seven pilot test participants. The participants were asked to rank the quality of their trip, the factors that influenced their trip quality, and the effectiveness of the video clips in terms of their ability to give the participants a feel for the traffic and roadway conditions they would experience if they were actually driving. An open conversation was then held for the researchers to get advice about how to improve the videotapes. The four most common factors that influenced the participants' trip quality were traffic volume, pavement quality, ability to maintain a constant speed, and the percentage of trucks. The participants' results were then compared to the density-based LOS calculated by the Highway Capacity Manual (HCM). Out of the nine video clips, three were rated the same as the HCM's density-based LOS. Five other clips were rated one letter grade lower than the HCM's density-based LOS ranking. The last video clip was rated three letter grades lower than the HCM's density-based LOS ranking. Some suggestions and recommendations given by the pilot test participants were to add sound, improve the video quality, videotape during different times of the day and varying weather conditions, and videotape more peak hour traffic conditions. Two complaints about this method were that it was hard to determine the roadway's pavement quality and it was hard for some participants to compare two-lane rural freeways to three-lane rural freeways. Out of the seven participants, three ranked this videotape method as very good. Another three ranked this method as good. One participant ranked this method as fair.
General Note: Title from title page of source document.
General Note: Document formatted into pages; contains 87 pages.
General Note: Includes vita.
General Note: Thesis (M.S.)--University of Florida, 2004.
General Note: Includes bibliographical references.
General Note: Text (Electronic thesis) in PDF format.

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 003090274
System ID: UFE0003401:00001

Permanent Link: http://ufdc.ufl.edu/UFE0003401/00001

Material Information

Title: An Exploratory data collection approach for the assessment of level of service from a traveler's perspective
Physical Description: Mixed Material
Language: English
Creator: Seager, Kimberly ( Dissertant )
Washburn, Scott S. ( Thesis advisor )
Courage, Dr. ( Reviewer )
Li, Tao ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2004
Copyright Date: 2004

Subjects

Subjects / Keywords: Civil and Coastal Engineering thesis, M.S
Dissertations, Academic -- UF -- Civil and Coastal Engineering

Notes

Abstract: Although the concept of level of service (LOS) for freeways is usually defined in terms of users' perceptions, there have been very few studies that have sought drivers, or passengers' views about what is important to them. The objective of this study was to develop and test a method by which driver perceptions of LOS for varying roadway conditions can be collected. The approach decided upon was to create video clips of rural freeway driving conditions from the driver's perspective. These video clips consisted of the front-view, the rear-view, the side-view, and the speedometer. Each video clip was between 2 to 2 1/2 minutes. In total, there were nine video clips. These video clips were then shown to seven pilot test participants. The participants were asked to rank the quality of their trip, the factors that influenced their trip quality, and the effectiveness of the video clips in terms of their ability to give the participants a feel for the traffic and roadway conditions they would experience if they were actually driving. An open conversation was then held for the researchers to get advice about how to improve the videotapes. The four most common factors that influenced the participants' trip quality were traffic volume, pavement quality, ability to maintain a constant speed, and the percentage of trucks. The participants' results were then compared to the density-based LOS calculated by the Highway Capacity Manual (HCM). Out of the nine video clips, three were rated the same as the HCM's density-based LOS. Five other clips were rated one letter grade lower than the HCM's density-based LOS ranking. The last video clip was rated three letter grades lower than the HCM's density-based LOS ranking. Some suggestions and recommendations given by the pilot test participants were to add sound, improve the video quality, videotape during different times of the day and varying weather conditions, and videotape more peak hour traffic conditions. Two complaints about this method were that it was hard to determine the roadway's pavement quality and it was hard for some participants to compare two-lane rural freeways to three-lane rural freeways. Out of the seven participants, three ranked this videotape method as very good. Another three ranked this method as good. One participant ranked this method as fair.
General Note: Title from title page of source document.
General Note: Document formatted into pages; contains 87 pages.
General Note: Includes vita.
General Note: Thesis (M.S.)--University of Florida, 2004.
General Note: Includes bibliographical references.
General Note: Text (Electronic thesis) in PDF format.

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 003090274
System ID: UFE0003401:00001


This item has the following downloads:


Full Text












AN EXPLORATORY DATA COLLECTION APPROACH FOR THE ASSESSMENT
OF LEVEL OF SERVICE FROM A TRAVELER' S PERSPECTIVE

















By

KIMBERLY SEAGER


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Kimberly Seager


































This document is dedicated to my two grandfathers who have passed away since I began
college.















ACKNOWLEDGMENTS

I would like to thank my committee chair and my committee chair members, Dr.

Washburn, Professor Courage, and Dr. Li. I would also like to thank Cody for his help

with all the equipment. I'd like to thank Brad Choi for going with me to collect data and

for convincing me to do a thesis. I would also like to thank David Kirshner and Luke

McLeod for helping me on this project.
















TABLE OF CONTENTS
Page

A C K N O W L E D G M E N T S ............................................................................................... iv

LIST OF TA BLE S ............... .............. ................ ......... .............. .. vii

LIST OF FIGURES ................... ...................... .. ................... viii

ABSTRACT .............. .......................................... ix

CHAPTER

1 IN TRODU CTION ............................................... ..... ..... .............. ..

G en eral B ackgrou n d .......... ................................................................. .. ............
Problem Statem ent .................... ................ ....... ............................. .............
Current Methodology Used to Define Service Measures.............................. 2
Applicable Service Measures for Urban and Rural Freeways ...........................3
R research Objectives and Tasks ........................................................ ............... 5
Organization of the Remainder of the Document .................................................5


2 L ITER A TU R E R E V IEW ................................................................ .....................6

Studies Pertaining to User Perceived LOS..................................... ............... 6
B background Survey .................. ...................................... .. ........ .... 17
Conclusions ................................. .. .... ........... ............. ........... 18
HCM Freeway Analysis Methodology Overview................................ 18
Urban and Rural Freeway Overview ........................................ ............... 18
Applicable Thresholds for Urban and Rural Freeways.................. ............19


3 R E SE A R C H A PPR O A C H ............................................................ .....................20

Alternative Approaches Evaluation ........................................ ...... ............... 20
V ideo D ata Collection M ethod ........................................ .......................... 22
V vehicle Instrum entation ............................................. ............................. 22
D ata C collection Site Selection....................................... .......................... 24
Video Data Collection Runs....... .............................................................27
V ideo C lip C om position........................................................... ............... 32
L oop D ata C collection ............................ ................................ .... ...... ...... 33


v









P ilo t T e st ....................................................................................................... 3 4
Pilot Test Procedure .................................... ...................................35
S u rv e y ....................................................................................................... 3 5

4 RESULT S ................ ................ ............... 38

Pilot Test Participant Feedback on Video Clips...............................................38
Pilot Test Participant Feedback on Survey Instrument.........................................39
Factors That Were Important in Evaluating Quality of Service.............................39
A Comparison of the HCM's LOS Evaluation to the LOS Ratings From the Pilot
T e st ................................................................. 4 0

5 CONCLUSIONS AND RECOMMENDATIONS.................................................43

C o n c lu sio n s ....................................................... ................ 4 3
V vehicle Instrum entation ........................................................... ............... 43
Determination of Data Collection Sites .......... .......... .............. ........43
Collecting Video Data ............. .... ................................... 44
Com position of V ideo Clips........... ................................................... .....44
D ata C ollection............................................. 44
P ilo t T e st ....................................................... 4 5
Recomm endations ......................... ..... .................. ......... 45


APPENDIX

A COM M ON ABBREV IA TION S ................................................... .....................46

B POSSIBLE SITE LOCATION TRAFFIC INFORMATION.............................47

C FINAL SITE LOCATION DETAILS........................................... .....................52

D SITE LOCATIONS AND DRIVING TIME ................................. ...............59

E LOOP DATA SUMMARY TABLES.................................. ....................... 63

F PILO T TEST SURVEY ................................................... ........ ............... .71

L IST O F R E F E R E N C E S ....................................................................... ... ................... 75

B IO G R A PH IC A L SK E TCH ..................................................................... ..................77
















LIST OF TABLES

Table page

3-1. Study Approach Alternatives M atrix ............................................ ............... 23

3-2. Final Site Location Information Table .......... ................................ ...............30

3-3. Loop D etector Traffic Inform ation ............................................... ............... 36

4-1. Pilot Test Participants Factors Affecting Trip Quality........................................39

4-2. Com prison of Results ..................................................... .......................... 42
















LIST OF FIGURES

Figure page

3-1. Test V ehicle................................................. 24

3-2. F ront-V iew C am era Set-up ........................................................... .....................25

3-3. Side-V iew Cam era Picture ............... .............. ........................................... 25

3-4. Speedometer Camera Picture ............................................................................25

3-5. Equipm ent Connection Schem atic ........................................ ....................... 25

3-6. Possible Site L ocations.................................................. ............................... 28

3-7. F final Site L location M ap ......................................... .. ............................ ..............29

3-8. Composite Video Screenshot ............................................................................33















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

AN EXPLORATORY DATA COLLECTION APPROACH FOR THE ASSESSMENT
OF LEVEL OF SERVICE FROM A TRAVELER' S PERSPECTIVE

By

Kimberly Seager

May 2004

Chair: Scott Washburn
Major Department: Civil and Coastal Engineering

Although the concept of level of service (LOS) for freeways is usually defined in

terms of users' perceptions, there have been very few studies that have sought drivers' or

passengers' views about what is important to them. The objective of this study was to

develop and test a method by which driver perceptions of LOS for varying roadway

conditions can be collected. The approach decided upon was to create video clips of rural

freeway driving conditions from the driver's perspective. These video clips consisted of

the front-view, the rear-view, the side-view, and the speedometer. Each video clip was

between 2 to 2 12 minutes. In total, there were nine video clips. These video clips were

then shown to seven pilot test participants. The participants were asked to rank the

quality of their trip, the factors that influenced their trip quality, and the effectiveness of

the video clips in terms of their ability to give the participants a feel for the traffic and

roadway conditions they would experience if they were actually driving. An open









conversation was then held for the researchers to get advice about how to improve the

videotapes.

The four most common factors that influenced the participants' trip quality were

traffic volume, pavement quality, ability to maintain a constant speed, and the percentage

of trucks. The participants' results were then compared to the density-based LOS

calculated by the Highway Capacity Manual (HCM). Out of the nine video clips, three

were rated the same as the HCM's density-based LOS. Five other clips were rated one

letter grade lower than the HCM's density-based LOS ranking. The last video clip was

rated three letter grades lower than the HCM's density-based LOS ranking.

Some suggestions and recommendations given by the pilot test participants were

to add sound, improve the video quality, videotape during different times of the day and

varying weather conditions, and videotape more peak hour traffic conditions. Two

complaints about this method were that it was hard to determine the roadway's pavement

quality and it was hard for some participants to compare two-lane rural freeways to three-

lane rural freeways. Out of the seven participants, three ranked this videotape method as

very good. Another three ranked this method as good. One participant ranked this

method as fair.














CHAPTER 1
INTRODUCTION

General Background

Transportation engineers are continually faced with identifying roadway

infrastructure improvements that will result in good benefit-to-cost ratios. Due to limited

financial resources, it is extremely important to make good decisions regarding selected

improvements. One of the most commonly used tools in the decision-making process is

the Highway Capacity Manual (HCM). The HCM is the basic reference and procedural

guide for traffic operations of analyses in the United States. It is one of the major

resources used to predict operational improvements resulting from a new or improved

facility. The HCM provides a systematic and consistent basis for analyzing the capacity

and level of service (LOS) for various highway and street facilities, including signalized

and unsignalized intersections, arterial streets, freeway segments, and highway segments.

The HCM is published by the Transportation Research Board (TRB) of the National

Academy of Sciences and is developed by the Highway Capacity and Quality of Service

(HCQS) committee. A new edition of the HCM was published in 2000, the first

complete revision of the HCM since 1985.

LOS assessment of a roadway facility has become a major foundation of the HCM.

The LOS concept is used in the HCM as a qualitative indicator of a traveler's trip quality

under specified roadway, traffic, and control conditions. In the 2000 version of the

HCM, LOS is described as a qualitative measure describing operational conditions within









a traffic stream, based on service measures such as speed and travel time, freedom to

maneuver, traffic interruptions, comfort, and convenience [1].

The 2000 HCM divides quality of service into six levels of service, LOS A LOS

F. LOS A represents excellent service and LOS F represents very poor service. If the

HCM analysis yields LOS A, the user knows clearly that the road is uncongested. This

facility has very low traffic volume and virtually no traffic congestion. At LOS C the

facility is in the mid-range of congestion. If the facility is at LOS E, it is approaching its

capacity, but still has undersaturated conditions. LOS F represents oversaturated

conditions. This occurs when the traffic demand exceeds the capacity of the facility.

Thus, the capacity of a roadway or intersection, as defined in the HCM, represents the

boundary between LOS E and F.

Problem Statement

Current Methodology Used to Define Service Measures

There is no commonly accepted quantitative procedure for setting LOS threshold

values (the service measure values that delineate one LOS value from another). The

criteria and breakpoints used to define LOS are based on the collective professional

judgment of the members of the HCQS committee of the TRB. Therefore, the LOS

threshold values represent the perspectives of transportation experts. The service

measure for each facility type has generally been selected based on two considerations.

The first consideration is that the service measure chosen should represent speed and

travel time, freedom to maneuver, traffic interruptions, and comfort and convenience in a

manner most appropriate to characterizing quality of service for the particular facility

type being analyzed. The second consideration is that the service measure chosen should









be sensitive to traffic flow rates so that the service measure characterizes the degree of

congestion of the facility [2].

However, the 1985 HCM described LOS as a qualitative measure that characterizes

operational conditions within a traffic stream and their perception by motorists and

passengers. The descriptions of individual levels of service characterize these conditions

in terms of such factors as speed and travel time, freedom to maneuver, traffic

interruptions, and comfort and convenience [3]. The quotation states that the perception

of driving conditions by drivers and passengers is an essential element in evaluating the

quality of service on a freeway facility. In addition, little has been done to compare the

expert opinions of the HCQS committee with the user's perceptions of LOS despite a

definition that suggests that it is driven by or meant to reflect user perception. When

these service measures were chosen the committee probably felt that they were highly

correlated with user perceptions, but they did not know for sure since users have rarely

been asked how they felt. Research on traveler perceptions of quality of service is

desirable so that the existing service measures can be validated (that the existing service

measures and thresholds are reflective of how roadway users actually perceive

conditions). This will assure transportation engineers and infrastructure investment

decision-makers that "improvements" identified as a result of LOS analyses will actually

be perceived as improvements by the traveling public. This will give the traveling public

more faith in the abilities of the transportation engineering community and those in

charge of managing the transportation system.

Applicable Service Measures for Urban and Rural Freeways

Rural freeways differ from urban freeways in that interchange spacing is much

greater for rural freeways, posted speed limits are higher for rural freeways, and rural









freeways have a much larger percentage of social and recreational trips while urban

freeways have a higher percentage of shopping and work trips, to name a few main

differences. Yet rural and urban freeways have the same service measure (density) and

thresholds for LOS, even though traveler perceptions and expectations are probably

different for rural freeway trips and urban freeway trips.

This belief is based on the premise that while urban freeway travelers experience

the full range of LOS values, rural freeway travelers rarely experience density conditions

that by current definition exceed LOS C. It is believed that travelers on rural freeways

usually expect low-density conditions, and even moderate amounts of congestion can

have a negative impact on a traveler's perceived quality of service being experienced [4].

If driver perceptions and expectations differ between rural and urban areas, the question

that arises is whether rural and urban freeways should have the same service measure

and/or thresholds.

The concept of different service measures and/or thresholds for a certain facility

type in areas with differing levels of development is nothing new. In fact, it is already

being used for a couple of facility types. Other roadway facility analysis methodologies

of the HCM have divided the facility type into different classes. For example, two-lane

highways have been divided into two classes. The two-lane highway analysis procedure

currently defines two classifications of highway by primary trip type served, with

differing service measures and thresholds. For the service measure in common between

the two classes (percent time spent following), the thresholds are different. Additionally,

the urban arterials analysis methodology uses one service measure, average travel speed,









but four different sets of thresholds, corresponding to four different arterial street

classifications.

Research Objectives and Tasks

The objective of this study was to develop and test a method by which driver

perceptions of LOS for varying roadway conditions can be collected. The method that

was chosen for this study was an in-vehicle video that would be shot from a driver's

perspective. This will be described in detail in the research approach chapter. It is hoped

that this method will provide the means to develop a procedure for estimating LOS on

rural freeways based on drivers' perceptions. Five major tasks were performed in the

development of this method. These tasks are as follows:

Determine how best to instrument the driving vehicle for the collection of rural
freeway data via video

Determine the rural freeway sites from which to collect video footage and traffic
operations data

Coordinate the simultaneous collection of loop detector data for the chosen sites

Produce the final video clips to be used for survey participant review

Pilot test the video scenes and LOS ranking survey


Organization of the Remainder of the Document

There are four remaining chapters in this document. Chapter 2 contains a summary

of relevant literature and an overview of the 2000 HCM freeway analysis methodology.

Chapter 3 contains the details of the research approach used for this study. Chapter 4

describes the results from the pilot test. The conclusions and recommendations section is

contained in Chapter 5.














CHAPTER 2
LITERATURE REVIEW

Although the HCM states that the concept of LOS should reflect traveler perception

there have been very few studies that have sought travelers' views about what factors are

important to them in assessing trip quality. A literature review was conducted to identify

previous research efforts into the specific issue of LOS perceptions by roadway users.

Studies Pertaining to User Perceived LOS

A study by Pecheux et al. [5] discusses how the right performance measures are not

being used to measure traveler perceptions and satisfaction. The authors felt this way

because there have not been many studies of drivers' perceptions done or written about.

Because there have been so few studies conducted to date in the area of traveler

perception of quality of service, it is safe to assume that the performance measures

currently in the HCM probably do not represent those most important to travelers.

Pecheux et al. feel that if the HCM really means that LOS is meant to reflect

drivers' and passengers' perceptions of the quality of service on a highway, then

engineers cannot choose the performance measures as they have in the past. Instead, the

public needs to decide what performance measures should be used to determine quality of

service. This paper concluded that more studies need to be done focusing on traveler

perceptions.

A study by Sutaria and Haynes [6] used a road-user opinion survey to evaluate the

current LOS methodology at signalized intersections. The survey involved depicting and

rating different traffic situations at a signalized intersection. Two types of film sequences









were used: a driver's view and an overhead view of an intersection. Over 300 drivers

rated randomly arranged films in terms of appropriate LOS. Field studies and the attitude

survey given to the drivers surveyed provided data for the development of two models.

According to the 1965 HCM, LOS is a qualitative measure of the effect of a

number of factors, which include speed, travel time, traffic interruption, freedom to

maneuver, safety, driving comfort, convenience, and operating costs [7]. Since there are

stops at intersections, speed cannot be the appropriate measure of LOS for arterials. That

led to the HCM establishing an operational index called load factor (LF). The HCM

defined this index as a ratio of the total number of green signal intervals that are fully

utilized by traffic during the peak hour to the total number of green intervals, for that

approach during the same period [7]. The LF was then used to determine the various

LOS at signalized intersections.

Sutaria and Haynes decided that their premise would be that the quality of flow at

any intersection should reflect the attitudes of the road users. Since there are different

levels of satisfaction regarding intersection operation, they felt that the opinions of a

group of road users could be used to establish a rational relationship between LF and

LOS. To do this would involve a compilation of all the road users' inclinations, feelings,

and degrees of satisfaction about the quality of service at an intersection. Drivers'

subjective ratings of quality of flow at a signalized intersection would represent their

attitudes or opinions, and LF, average individual delay (AID), or a saturation factor

would represent an objective rating.

The film segments chosen would represent various specific levels of service and

would be shown to the drivers for a duration of one or two signal cycles. The final film









prepared was a compellation from one or two representative film segments for each LOS.

The final film also included both types of view, and the order of the segments in each

type was determined on a random basis.

The survey consisted of two parts: a group attitude survey and the film survey. A

user-response survey was decided upon because such an approach had been successfully

utilized previously and showed that it could provide meaningful results. A questionnaire

was developed for the group attitude survey. Drivers were asked to answer the first part

of the questionnaire before they viewed the segments of film. The first part inquired

about sex, age, driving experience, education, and type of roadway most often driven on.

They were then requested to rate their opinions of five factors regarding the quality of

service at an intersection: delay, number of stops, traffic congestion, number of trucks

and buses, and difficulty of lane changing. Then the first part of the film was presented

to them segment by segment. Each segment consisted essentially of a driver's view of

approaching, waiting, and passing through the intersection. After each segment the group

was requested to rate the traffic operations with regard to LOS provided. After the

presentation of the film, the drivers were requested to again rate the five factors regarding

quality of flow at the intersection. This was done in order to find out if the films

influenced their initial opinions. In all, 310 respondents participated in the survey at

several sessions.

The results from the group attitude survey indicated that delay was considered the

most important factor both before and after viewing the film segments. The other four

factors changed in rank before and after viewing the films. Based on the results from the

film survey, the hypothesis that LF is a better predictor of LOS than AID was not true.









This strongly supports the argument that the definition of LOS as a function of LF was

not adequate. This lead to a new relationship among AID, LF, volume to capacity (v/c)

ratio, and LOS based on user perception. This study was one of the first that asked users

how they felt about traffic conditions. The results from this study changed the measure

used by the HCM to assess LOS at signalized intersections.

An additional study conducted by Pecheux et al. [8] addresses methodological

issues faced in the development of a study to assess user perceived LOS at signalized

intersections. This study had two main objectives. The first objective was to determine

the appropriateness of the current LOS methodology by seeking users' opinions. The

second objective was to determine the factors affecting users' LOS perceptions at

signalized intersections.

Although the concept of LOS is meant in part to reflect the operational conditions

as perceived by motorists, the HCM LOS thresholds for signalized intersections were not

based on studies of driver perceptions. The HCM criteria were created from observed

field delays. Thus, each LOS represents different delay conditions, but not necessarily

the delay that motorists perceived. The HCM specifies average control delay as the

measure of effectiveness for signalized intersection LOS analysis, but it is unlikely that

delay is the only factor that influences user perception of quality of service. The HCM

uses the same delay thresholds that were established 15 years ago. Thus, travel

conditions that would have been viewed as intolerable in the 1960s are considered normal

by today's motorists, especially commuters.

Methods of data collection evaluated for this project included: on-the-road field

studies, controlled test-track studies, and controlled laboratory studies. The researchers









decided that subjects would be shown videos as though they were the driver. This method

would allow for multiple subjects to be run simultaneously and give the researchers

control over the experimental conditions.

A questionnaire with three parts was designed to gather detailed subject

information. The first part was designed to explore the subjects' attitudes about driving

in certain situations. The second part of the questionnaire was designed to explore

personal characteristics of the subjects. Socio-demographic information was obtained in

the third part of the questionnaire. Subjects were not persuaded to use delay for rating

LOS, they were simply told to rate the quality of service provided by the traffic signal.

The results of the survey show that subjects were more tolerant of delays than the

HCM would suggest, and their LOS ratings tended to be similar for delays associated

with LOS A and B and LOS C and D. The results also showed that subjects' delay

estimates were fairly accurate, but widely variable, as were their perceptions of LOS.

Although a few subjects used delay as the only criterion when rating LOS, most subjects

considered more than just delay. Fifteen factors were identified by the subjects as

influential in their LOS ratings. They were as follows:

* delay
* size of intersection
* traffic signal efficiency
* pavement quality
* arrows/lanes for turning vehicles
* queue length
* visibility of traffic signals from queue
* traffic mix
* clear/legible signs and road markings
* location
* geometric design of intersection
* scenery/aesthetics
* leading left-turn phasing scheme
* presence of pedestrians









visual clutter/distractions


These results suggest that users' perceptions of LOS are sensitive to factors other

than delay. The authors suggest in future experiments that the most important factors

influencing users' perceptions first be identified and then controlled in the experimental

design.

A study by Hall et al. [9] reports on the results of focus group sessions in which a

group of commuters discussed their views about determinants of the freeway quality of

service they experienced. The researchers tried to figure out what aspects of freeway

travel are important to motorists. Hall et al. concluded that total travel time is the most

important determinant for commuters, but a number of other aspects of the trip also

mattered. These aspects included safety, traveler information, and maneuverability

(density).

This research used focus groups in order to determine what aspects of freeway

travel are most important to motorists. There were two focus groups, one with five

people, and the other with seven members. Focus group members were all university

faculty members, in a wide range of departments. The total population of drivers

included non-commuters as well as commuters, truck drivers as well as drivers of

passenger cars, and a wide range of backgrounds and education. More men than women

participated in this study. These people were selected to ensure that all of the participants

traveled the same stretch of freeway, so that they all knew about the situations being

discussed, and had relatively similar experiences. The research team felt that this would

provide the most productive context for an exploratory analysis of users' perceptions.









The focus groups lasted an average of one and a half hours. In the focus groups,

perceptions of trip quality and the factors that influence trip quality were investigated, as

well as the factors that influence changes in perception from trip to trip. The facilitator

guided the focus group participants using an interview topic checklist containing a list of

the topics relevant to the research, and a series of semi-structured, open-ended questions

designed to draw out information on these topics. Finally, the researchers asked about

the differences between being a driver and being a passenger. This was included in the

topics discussed because the HCM states that the perceptions of both drivers and

passengers are important. Preliminary results were circulated to research participants to

help assess the credibility of the researchers' interpretations of the experiences discussed.

Responses indicate that the focus group participants agree with the researchers'

interpretations of their discussions.

The results from the focus groups identified four primary factors that were

important to urban freeway motorists. These factors were travel time, density, safety, and

traveler information. Travel time is the first thing respondents used to describe the

quality of a particular trip. How long it takes to get from where they are to where they

need to be was the most important measure of the quality of that particular trip. Having

time constraints on their arrival greatly increases the stress involved in the trip and with it

the perceived quality of the trip. The participants also felt time spent commuting was lost

time as far as they were concerned. An indicator of the importance of travel time was

that when researchers asked about trips that were less than desirable the first item that

came up in response was construction. Construction was mentioned because it tended to

slow traffic down. The other indicator was the length some people went to in order to









keep the travel time relatively short. An example of this would be leaving the house for

work at 4 a.m. instead of 6 a.m.

A number of other issues were mentioned in the focus group: density, accidents,

travel information, driver civility (or incivility), the use of photo radar, and weather. The

way the focus groups felt about density was best summarized by the following comment.

Even if there's an extraordinary volume of traffic, as long as it's clicking along, even if

there's cars all around you, and trucks and everything, as long as everybody's going then

it's just a question of time [9]. Participants were concerned about accidents not only in

terms of congestion, but also because of the risk to their personal safety. Having

adequate information about what was happening to traffic while they were on the road

was also important to participants. Driver civility and politeness were mentioned in

reference to the notion of lane etiquette (i.e. using different lanes for different speeds and

to pass).

The responses of both focus groups made it clear that people have a very different

perspective on the trip when they are passengers as compared to drivers. The focus group

participants felt that the tension and sense of responsibility are much less as a passenger

than as a driver. Also, there is not the same degree of concern about the travel time. This

paper was helpful because focus groups were one of the options considered in the

alternative approaches evaluation. This paper also showed that density is not the only

factor considered by drivers.

A paper by Hostovsky and Hall [10] deals with the perceptions of tractor-trailer

drivers regarding the quality of service on freeways. This was accomplished by a focus

group with professional tractor-trailer drivers. Because of the different performance









characteristics between automobiles and tractor trailers, their reasons for being on the

road, and the amount of time spent on the road, it is reasonable to expect that different

factors will affect truck drivers' perceptions of freeway quality of service. The paper also

compares and contrasts tractor-trailer drivers' perceptions with factors that are important

to freeway automotive commuters.

The focus group was held on November 15, 2001 and consisted of seven tractor-

trailer drivers. The session lasted for one and a half hours. The focus group was

structured the same way as described in Hall's previous paper [5]. The researchers

concluded that freeway conditions in general were the most frequently mentioned factors.

The three variables that were used to describe traffic conditions were: travel time (or

speed), traffic density (or maneuverability), and traffic flow. The most significant

finding was that it is traffic flow that matters to tractor-trailer drivers, not density. Other

important items that affected the perceived quality of service included weather, attitudes

toward other drivers, and road rage.

Some of the freeway conditions discussed included the condition of the road

surface, lane restrictions, lane width, lane markings, and signage. These factors did not

come up in the previous focus groups with freeway automobile commuters. The tractor-

trailer drivers contended that trucks have been getting longer and wider at the same time

lanes are getting narrower. The condition of the road surface was not the only concern to

tractor-trailer drivers, but traffic conditions were also important. A steady traffic flow

within an acceptable range was more important to them than actual traffic density.

It was clear from the focus group discussion that although tractor trailer drivers are

concerned with some of the same characteristics of a freeway journey as are automobile









commuters, they also have some different concerns. Urban freeway commuters' primary

concern was travel time. Steadiness of the traffic flow was more important in terms of

trucker's perception of the quality of service.

Three other issues were also important to these truckers for their overall opinion

on trip quality. The first was simply the physical condition of the roadways. The second

was maneuverability. Urban commuters seemed content to stay in one lane. Meaning,

maneuverability was not important in the urban context due to higher traffic density. The

third trucking issue was the safety of the trip. Safety was also an important concern with

urban commuters, although the nature of their concerns differed. This paper was

informative because it went on to compare and contrast tractor-trailer drivers' perceptions

with factors that are important to freeway automotive commuters.

In a study by Hostovsky et al. [11] focus group participants identified three themes

important for rural freeways LOS. These three themes included the ease of rural driving

(i.e. low density), predictable travel time and steady travel speed, and ample

maneuverability on rural freeways. There was also concern about safety related problems

on rural freeways, such as problems associated with the isolated nature of rural freeways

in association with weather conditions. Secondary rural freeway themes included

aesthetics, speeding, the discomforting presence of trucks, and the need for better

commuter information. This is an important paper because it showed that rural freeway

driver's consider other factor besides density in rating the quality of service of their trip.

A paper by Nakamura et al. [12] summarizes a study performed along the Tomei

Expressway in Japan on November 27, 1998. In this study, traffic flow conditions along

a section of a basic intercity motorway were evaluated from the driver's viewpoint. The









study consisted of a field driving survey that was intended to collect data on the degree of

driver's satisfaction under various uncongested traffic flow conditions along a rural

motorway. The major concern of this study was to quantitatively analyze the

interrelationship between traffic flow conditions (traffic volume, speed and lane

utilization ratio), drivers' perceptions (utility and degree of satisfaction) and driving

behaviors (lane changing, adjustment of acceleration and adjustment of spacing) [12].

The objective of the field driving survey was to collect data on the perception of

drivers and to observe driving behavior under various uncongested traffic flow

conditions. Twenty-two subject vehicles drove themselves from the on-ramp to the off-

ramp and back along the Tomei Expressway. A video camera was mounted on the

navigator's seat of each vehicle to have the front views recorded. After every trip the

subject driver was asked to answer a survey about the traffic flow conditions. A total of

105 surveys were collected. Each traffic flow condition was evaluated using a scale from

one to five. The data collected for driving behavior was the number of lane changes,

elapsed travel time by lane, and percentage of time spent following another vehicle.

It was found that the factor that most strongly affected the degree of driver's

satisfaction was the traffic flow rate. The number of lane changing, the percentage of

time spent following, and the driving experience also affected the evaluation of the traffic

condition. In addition, setting the LOS based on the average degree of driver's

satisfaction was tried and was compared with the conventional LOS measures. The

results suggest that the current traffic conditions on Japanese motorways hardly satisfy

drivers. Meaning, if highways are designed employing the volume to capacity (v/c) ratio

that is set only from the driver's degree of satisfaction it would take an enormous









monetary investment. This paper was helpful not only because it seeks users' opinions

but because having participants drive along rural freeways was one of the options

considered in the alternative approaches evaluation. A few things to consider about this

study are that this study was performed in Japan. Also, it collected 105 surveys from

only twenty-two subjects.

In a study initiated by the Wisconsin Department of Transportation (WDOT) [13]

video segments of traffic operations at selected points along a highway were shown to

subjects to try and determine what factors were important to motorists. Survey forms

were also used to gather data and correlate it with field measurements of operational

measures of effectiveness' (MOE). In addition, qualitative input was sought from road

users, using focus groups to gain further insights. Survey respondents were shown 50-

second video clips for six different four-lane rural freeways. Field-measurements were

made to determine volume and speed for the video segments shown to the subjects.

Density was determined by calculation. The results showed that density and speed were

both found to be significantly related to the subjective rating of overall quality of flow.

This was a helpful paper because it performed a similar study (by showing video clips to

participants and asking their opinion of the roadway conditions) to the type of research

that will be performed.

Background Survey

A study conducted at the University of Florida [4] was performed hoping to obtain

preliminary information about what factors are important to driver's when evaluating the

quality of their trip on a rural freeway. After consideration of several potential methods

for obtaining direct traveler input (i.e., focus groups, post simulation/video review,

interview, etc.), it was decided to use a survey-based approach. One of the intents of this









survey-based approach was to help guide a more focused follow-on study that might

make use of an alternate method that will only allow for a relatively small sample size.

Two-hundred and thirty-three travelers were surveyed. The most important factors to

drivers were decided on by the percentage of time they were ranked first, second, or third

by drivers. The surveys concluded that the most important factor to drivers was the

ability to consistently maintain their desired travel speed. This was in the top three

64.3% of the time. The second ranked factor was the ability to change lanes and pass

other vehicles easily. This was part of the top three responses 33.3% of the time. The

third most important factor to drivers was the ability to travel at a speed no less than the

posted speed limit. This was in the top three 33.0% of the time. These results show that

although density is important to drivers it is not the most important factor, driver's

consider other factors besides density in the quality of service of their trip.

Conclusions

A recent survey of customer-oriented practices of highway agencies pointed out

that a customer-oriented perspective not only increases the quality of performance but

also helps build and sustain necessary public support for transportation programs [14].

This literature review has shown while it may not be possible to evaluate how drivers

perceive the quality of their trip with extreme precision and confidence, it is still possible

to arrive at useful conclusions which can help the decision makers more closely reflect

user perceived quality of service.

HCM Freeway Analysis Methodology Overview

Urban and Rural Freeway Overview

The HCM defines a freeway as a multilane, divided highway with a minimum of

two lanes for the exclusive use of traffic in each direction and full control of access









without traffic interruption [1]. The service measure used to determine LOS for freeways

is density (passenger cars per mile per lane). Density is a function of both speed and flow

rate, which incorporates consideration of freedom to maneuver, driver comfort and

convenience.

Applicable Thresholds for Urban and Rural Freeways

LOS has evolved throughout each revision of the HCM. The original HCM used a

basic three-point scale to define level of capacity. In 1963 level of capacity was

abandoned for the LOS concept. In 1965 a six point LOS definition was introduced in

the HCM, LOS A LOS F. In 1985 the six point LOS scale was redefined to include

traffic density (vehicles per unit length of roadway) as the measure for defining LOS for

basic freeway sections. This is still used today.

Not only do rural and urban freeways have the same service measure, density, but

they also have the same limits for LOS. LOS thresholds for a basic freeway segment are

summarized below.

LOS Density Range (passenger cars/mile/lane) (pc/mi/ln)

A 0-11

B > 11-18

C > 18-26

D > 26-35

E > 35-45

F >45














CHAPTER 3
RESEARCH APPROACH

Alternative Approaches Evaluation

The objective of this study was to develop and test a method by which driver

perceptions of LOS for varying roadway conditions can be collected. It is hoped that this

method will provide the means to develop a procedure that will lead to specific

recommendations for rural freeway LOS criteria and thresholds. An alternative

approaches evaluation was created to weigh the pros and cons for methods of collecting

driver's input about their driving experiences on rural freeways. Methods of data

collection evaluated for this project included: focus groups, in-field surveys, videos from

an overhead perspective, videos from a driver's perspective, a driving simulator, driving

test participants on rural freeways, and having test participants drive themselves on rural

freeways. These are now described in more detail.

* Focus Group: This method would consist of recruiting individuals to participate in
group discussions with a facilitator to guide the discussion in a constructive manner
such that the participants reveal the most important factors to them in evaluating
quality of service.

* In-Field Survey: This approach would consist of developing a more focused
survey based on the results from the University of Florida's study [4]. The survey
would solicit driver opinions about various factors related to the perception of the
quality of their trip on a rural freeway.

* Videos From an Overhead Perspective: This method would consist of having test
participants review pre-recorded traffic scenes from an overhead perspective and
then answer survey questions related to the observed video scenes.

* Videos From a Driver's Perspective: This approach would consist of test
participants reviewing pre-recorded traffic scenes from a driver's perspective and
then answering survey questions related to the observed video scenes.









* Driving Simulator: This approach would consist of having test participants drive
sections of a simulated rural freeway in which traffic and roadway conditions are
systematically varied. Specific LOS thresholds would be determined from
observed driving behavior and questionnaires administered either during or after
the simulation exercises.

* Driving Test Participants Around on Rural Freeways: This approach would consist
of having test participants ride in a vehicle driven by a research official, during
which the participants would fill out a trip quality questionnaire, with responses
being dependent on the given roadway and traffic conditions being experienced at
that time.

* Recruiting Test Participants to Drive on Rural Freeways: This method would
consist of having test subjects drive vehicles in the traffic stream and provide the
researchers with real-time feedback either instrumented or verbal.

A matrix was developed for these seven different methods of data collection. This

matrix is shown in Table 3-1. The matrix had nine different criteria on which each

approach was judged. For each criterion, each approach was evaluated with a low,

medium, high, or not applicable (NA) ranking, depending on how the criterion related to

the approach. The nine criteria chosen for this matrix were: number of test subjects at

once, control over experimental conditions, liability issues, responses while conditions

still fresh in mind, supplemental data collection, driver's perspective, concern of motion

sickness, ability to relate to actual driving situation, and being directly in control of the

driving situation. These criteria are now described in more detail.

* Multiple Test Subjects at Once: This criterion evaluated the number of test
subjects that could participate at one time.

* Control Over Experimental Conditions: This evaluated whether the researchers
would be able to control the experimental conditions (i.e. density, percentage of
trucks) at the time the subjects participated in the study.

* Liability Issues: This criterion dealt with the liability concerns for the test
participants and/or researchers at the time the study was prepared and conducted.

* Responses While Conditions Still Fresh in Mind: This criterion evaluated if the
responses from the subjects would occur during, right after, or a considerable
amount of time after a rural freeway trip.









* Supplemental Data Collection: This criterion addressed if supplemental data
collection (i.e. density, percentage of trucks) would be necessary and if so to what
extent.

* Driver's Perspective: This criterion evaluated if the approach would give the
subject a perspective as if he/she were the driver.

* Concern of Motion Sickness: This criterion addressed the possibility of
participants getting motion sickness during the study.

* Ability to Relate to Actual Driving Situation: This criterion evaluated the subject's
ability to relate this approach to a real world driving situation.

* Being Directly in Control of the Driving Situation: This criterion evaluated if the
participant was able to control the movements made during the study. It explores if
the participants were able to drive how they would normally drive on the rural
freeway.

The in-vehicle video approach was thought to offer the best compromise between

a realistic driving experience, study efficiency, and liability concerns. As the literature

review has shown, video studies in laboratory situations have been successfully used in

traffic perception studies for at least 40 years.

Video Data Collection Method

Since it was concluded to show test participants videos as if they were the driver, a

data collection method was developed for this task. The development of this method

consisted of six parts: vehicle instrumentation, determination of data collection sites,

collecting video data, composition of video clips, data collection, and a pilot test.

Vehicle Instrumentation

Again, the objective was to create video clips of rural freeway driving conditions

from the driver's perspective. It was decided to capture the following fields-of-view

(FOV) from within the vehicle:

* Front windshield and rear-view mirror
* Side-view mirror
* Speedometer









Table 3-1: Study Approach Alternatives Matrix

~j U- U =(.2 _t .=






Focus Group Med NA Low Low Low Low NA Low NA


In-Field Surveys High Low Low High Low High NA Med High

Videos from an Overhead
es veMed High Med High High Low Low Low Low
Perspective
Videos from a Driver's
Med High Med High High Med Med Med Low
Perspective

Driving Simulator Med High Low High Low Med High Med High

Driving Test Participants Low Low High High High Med Med High Low
on Rural Freeways

Having Test Participants
ving et Partiiats Low Low High High High High Low High High
Drive on Rural Freeways









The vehicle used for instrumentation was a Chevrolet Venture mini-van. The

vehicle is shown in Figure 3-1. A total of three video cameras were used. Each video

camera was attached to a VCR. A video monitor was attached to each VCR such that a

visual confirmation could be made of each

camera's FOV. A microphone was also

S connected to the audio input of one of the VCRs.

Portable 12-volt batteries were used as the power

source for all the equipment. The front-view

camera was set up on a pole between the two front
Figure 3-1: Test Vehicle
seats. It was attached to the driver's armrest.

This is shown in Figure 3-2. The side-view camera was attached to a pole between the

driver's seat and door. This is shown in Figure 3-3. The third camera was positioned on

the dashboard to capture the speedometer. Duct tape was used to cover the instrument

panel to help reduce the glare. This is shown in Figure 3-4. An equipment connection

schematic is shown in Figure 3-5.

Data Collection Site Selection

The rural freeway sites used were all within Florida. It was desired to obtain actual

traffic conditions at the time the video data collection runs were made. Thus, one of the

requirements in site selection was the presence of inductance loop detectors on the

segment. The Florida Department of Transportation (FDOT) maintains an extensive

network of inductance loop detectors (ILD) on the Florida Intrastate Highway System

(FIHS). There are over 7,500 detector stations on the state's highways. There are two

types of ILD stations, telemetered and portable. The telemetered stations were preferred



























figure _-z: rront-View figure 3-3: alde-view
Camera Set-up Camera Picture


Figure 3-4: Speedometer
Camera Picture


Figure 3-5: Equipment Connection Schematic

because they record data on a continuous basis (365 days a year) and are permanent

stations. The portable stations require a traffic data recorder to be installed in a roadside

cabinet (adjacent to the ILD station). Each ILD station has a unique identification

number assigned to it. The data collected at the ILD stations is compiled every year and

made available on the Florida Traffic Information (FTI) CD [15]. Some other









information that the FTI CD provides is the number of lanes in each direction, the site

type (telemetered or portable), a site description, the Annual Average Daily Traffic

(AADT) information, the percentage of trucks, and the peak hour in each direction.

All the freeway locations with telemetered loop detectors in the state of Florida

are shown in Figure 3-6. They are also listed in Appendix B with their site descriptions

and AADT information. Not all of these stations are located on rural freeway segments,

so Geographic Information System (GIS) software was used to help determine suitable

locations. Eight specific site locations were then decided upon. These eight locations

can be seen in Figure 3-7. A more detailed breakdown of the traffic conditions at these

sites can be found in Table 3-2. Appendix C provides graphical information for

these final eight site locations.

Both two-lane and three-lane rural freeway sections were selected. South Florida

was generally not considered due to limited number of rural locations and long distance

from Gainesville. Also, the Panhandle was not considered due to driving distance

reasons. The Turnpike site was chosen for its rolling terrain. The characteristics

possessed by these eight site locations address the top four concerns to motorists

according to the survey conducted at the University of Florida [4]. These four items

addressed are: ability to maintain travel speed, ability to change lanes, ability to exceed

the posted speed limit, and the percentage of heavy vehicles.

It was desired to collect a good range of traffic and roadway conditions. As many

different LOS values between A and E as possible was desired. This required the

experimenters to try to find low, medium, and heavy traffic conditions for these rural

freeways. The FTI CD provided the times for peak hour traffic. These times can be seen









in Table 3-2. In order to determine the LOS of the rural freeway at the time the roadway

was filmed, data that was recorded by the loop detectors would need to be obtained from

the FDOT. The FDOT normally has the loop detectors programmed to report the traffic

conditions in one-hour time intervals. Since traffic information for the exact time the

loop detector was passed would be preferred, the FDOT reprogrammed their loop

detectors to report the traffic data in five-minute intervals.

Video Data Collection Runs

After coordination on the reprogramming of the telemetered data collection stations

for five-minute intervals, the actual data collection driving runs were made. Three

different drivers made driving runs. All used the same basic driving strategy, which

consisted of attempting to drive five mi/hr over the posted speed limit (that is, 75 mi/hr)

and generally staying in the right lane unless it was necessary to move to another lane to

stay at the desired travel speed (such as a passing maneuver).

Each rural freeway was videotaped for approximately ten miles before and after

the loop detector to ensure consistent traffic conditions. The microphone was used

during the experiment to record time, date, and location information on the audio track of

one of the videotapes. Each time a loop detector was passed, the route, site number,

direction, and time of day was spoken into the microphone.

The first day of filming occurred on Monday, November 3, 2003. The sites filmed

were along 1-75 and 1-10 (site numbers 290320, 290269, and 299936). The segment

containing the 1-75 loop detector (site number 290320) was videotaped in the northbound

direction at 3:24 p.m., 4:24 p.m., and 5:33 p.m. The same segment was videotaped in the

southbound direction at 4:18 p.m. and 5:17 p.m. The segment containing the furthest

west loop detector along I-10 (site number 290269) was videotaped in the eastbound















Possible Site Locations


Legend
S Telemetered Sites
Freeways
N


s

rLLLI] Miles
0 1530 60 90 120


t. **







*




..; ,







*^ I
1,


Created 10/2003


Figure 3-6: Possible Site Locations















Final Site Locations














I, .


N
F L, ,t
















S

FLFI Miles
0 1530 60 90 120
Created 10/2003
Created 10/2003


Figure 3-7: Final Site Location Map












Table 3-2: Final Site Location Information Table


# of
Lanes Truck Peak Hour Peak Hour Peak Hour
Site Each Lane Truck Truck Volume Direction 1: Direction 2: Mile
# Route Number Direction Restraint AADT % AADT (AADT*K*D) N,E S,W Post
302-
1 1-10 290269 2 No 5,228 25.89 20,192 1,661 6 p.m. 5 p.m. 301
312-
2 1-10 299936 2 No 5,502 27.12 20,287 1,446 6 p.m. 5 p.m. 311
428-
3 1-75 290320 3 Yes 11,685 25.89 45,131 3,347 4 p.m. 5 p.m. 429
429
288-
4 1-95 730292 2 No 5,072 8.94 56,736 3,213 5 p.m. 4 p.m. 289

1 p.m. &
5 1-75 269904 3 Yes 12,823 21.85 58,687 4,182 4 p.m. 376
4 p.m.
325-
6 1-75 189920 2 No 7,016 17.99 39,000 2,356 Unavailable Unavailable 32
324
281
7 Turnpike 970428 2 No 4,186 12.34 33,926 1,950 6 p.m. 6 p.m.
278-
8 1-75 140190 2 No 13,688 19.36 70,702 3,569 6 p.m. 8 a.m. 27
279









direction at 3:30 p.m., 4:32 p.m., and 5:40 p.m. This segment was videotaped in the

westbound direction at 4:09 p.m. and 5:08 p.m. The last site recorded on November 3rd

was the segment containing the east loop detector on 1-10 (site number 299936). This

segment was videotaped in the eastbound direction at 3:40 p.m., 4:39 p.m., and 5:48 p.m.

This segment was videotaped in the westbound direction at 4:02 p.m., 5:01 p.m., and 6:09

p.m.

The second day of filming was to record the loop detector on 1-95 just south of

Palm Coast (site number 730292). This segment was videotaped on Tuesday, November

4, 2003. The segment was videotaped in the northbound direction at 12:55 p.m., 1:21

p.m., 1:46 p.m., 5:04 p.m., and 5:35 p.m. The southbound direction of this segment was

videotaped at 1:13 p.m., 1:38 p.m., 4:56 p.m., and 5:22 p.m.

The third day of filming was to capture the Tampa and Orlando site locations. This

occurred on Wednesday, November 5, 2003. The segments videotaped were loop

detectors along 1-75 by Micanopy (site number 269904), by Wildwood (site number

189920), by County Road 54 (site number 140190), and along the Turnpike (site number

970428). The first segment, by Micanopy (site number 269904), was videotaped heading

northbound at 11:05 a.m. and 11:20 a.m. This segment was videotaped in the southbound

direction at 11:01 a.m., 11:16 a.m., and 11:32 a.m. The segment by Wildwood (site

number 189920) was videotaped heading northbound at 12:19 p.m. and southbound at

12:13 p.m. and 12:29 p.m. The segment south of County Road 54 was videotaped

heading northbound at 1:58 p.m. and 2:17 p.m. and southbound at 2:11 p.m. The

segment along the Turnpike (site number 970428) was videotaped heading northbound at









4:18 p.m. and 4:50 p.m. The southbound direction of this segment was videotaped at

3:21 p.m. and 4:33 p.m. Filming this day was cut short due to weather conditions.

The fourth, fifth, and sixth days of filming recorded the same sites as the third day

did. The fourth day was Friday, November 7, 2003. The segment near Micanopy (site

number 269904) was videotaped in the northbound direction at 11:55 a.m., 12:13 p.m.,

and 4:48 p.m. This segment was videotaped in the southbound direction at 12:09 p.m.

and 12:28 p.m. The segment near Wildwood (site number 189920) was videotaped

heading northbound at 1:16 p.m. and 1:34 p.m. and it was videotaped southbound at 1:09

p.m. and 1:28 p.m. The Turnpike segment (site number 970428) was videotaped heading

northbound at 3:20 p.m. and southbound at 2:48 p.m. and 3:28 p.m. The fifth day of

filming was Sunday, November 9, 2003. The segment by County Road 54 (site number

140190) was videotaped in the northbound direction at 3:00 p.m. and the segment near

Wildwood (site number 189920) was videotaped heading northbound at 3:35 p.m. The

sixth day of filming occurred on Friday, November 21, 2003. The segment videotaped

on this day was the one near County Road 54 (site number 140190). This segment was

videotaped in the northbound direction at 6:37 a.m., 6:48 a.m., 6:59 a.m., 7:12 a.m., 2:18

p.m., 2:28 p.m., and 2:39 p.m. This segment was videotaped in the southbound direction

at 6:41 a.m., 6:52 a.m., 7:04 a.m., 7:15 a.m., 2:11 p.m., 2:22 p.m., and 2:32 p.m. A table

of these site locations and driving times can be seen in Appendix D.

Video Clip Composition

The survey participants were shown a video that consists of a single composite

scene of the three views recorded with the video cameras in the mini-van. The front-view

video was the main part of the final video. The front-view video showed the driver's

front-view with the rear-view mirror in the upper middle. The side-view mirror video









was placed in the upper left corner. The speedometer video was shown to the right of the

rear-view mirror. The video editing software program, Adobe Premiere [16], was used to

create these composite videos. A composite video screenshot is shown in Figure 3-8.

Each video was approximately 2 to 212 minutes in length. This length was chosen

because the researchers felt that it was a good compromise between consistent traffic

conditions experienced on the rural freeway at the time of filming and the survey

participants' attention span.


















Figure 3-8: Composite Video Screenshot

Loop Data Collection

In order to determine the LOS of the traffic conditions when the video runs were

made based upon the HCM's density criterion, loop data was simultaneously recorded

with the video segments. The loop data collected included speed, volume, and vehicle

classification. The telemetered loop detectors were collecting traffic information while

the rural freeways were being filmed. The FDOT was able to provide this traffic

information to the researchers in five-minute intervals. The information provided by the

FDOT included a traffic count file, a classification file, and a speed file for the eight









telemetered site locations used in this study. A summary table was created to show the

traffic conditions at the times the loop detectors were passed. This table can be seen in

Appendix E. From this data the density and heavy vehicle percentage for the site location

in the five-minute period in which the segment containing the loop detector was

videotaped was calculated. It needs to be mentioned that the conditions seen in the video

scene could still vary substantially from the average five-minute conditions calculated

from the collected ILD data. Based on these calculations, nine 2 to 212 minute video

clips were chosen to be used for the pilot test. The criteria possessed by the nine video

clips selected were:

1. Low density, two lane
2. Low density, two lane, rolling terrain
3. Low/Medium density, three lane
4. Medium density, two lane
5. Medium density, two lane, rolling terrain
6. High density, two lane
7. Low heavy vehicle percentage, three lane
8. Medium heavy vehicle percentage, two lane
9. High heavy vehicle percentage, two lane

The traffic information, including the LOS based on the HCM's density criteria, for these

nine video clips, is shown in Table 3-2.

Pilot Test

The main purpose of this pilot test was to see if the developed method could be

successful at obtaining useful driver perception data on rural freeway LOS. The

researchers wanted to determine if the pilot test participants felt they were able to relate

to the videos from a driver's perspective. The researchers also wanted to find out if this

was a realistic method and what improvements could be made to make this study more

accurate.









Pilot test procedure

Survey participants for the pilot test were University of Florida civil engineering

undergraduate students. Since these participants had just completed the introductory

transportation engineering course, they were familiar with the freeway LOS

methodology. Because of the video format, multiple test participants could be "run"

simultaneously while allowing the researchers control over the test conditions. As many

participants as the room allowed to have a clear view of the screen could be tested at the

same time. The video screen was placed at approximately sitting eye level so the

participants were looking at the screen as if they were looking out of a car's windshield.

Seven students participated in the pilot test. The subjects were shown nine 2 to 212

minute videos. Before viewing the videos, test participants were given only the basic

instruction to imagine themselves driving the vehicle in these conditions, and to make

their trip quality ranking based on how they would drive in those conditions, not

necessarily how the driver of the instrumented vehicle was driving. The conditions

shown by the video clips were representative of an extended rural freeway trip. After

each video, the participants were asked to fill out a survey.

Survey

A test survey was developed for use with the pilot test. The main objective was to

identify the general suitability of the survey and any potential improvements. Test

participants gave a trip quality ranking after viewing each video. Subjects were not

persuaded to use density as the criteria for rating the trip quality. They were simply told

to just rate the quality of the trip. Trip quality was evaluated using six classifications:

excellent, very good, good, fair, poor, and very poor. These classifications were generally










Table 3-2: Loop Detector Traffic Information


5 Total 24
Clip Date Total Avg. min 24 Hr 24 Hr 24 Hr Hr 24 Hr
No. (2003) Site Time Dir. Volume Speed vol veh/hr/ln Density Vol Buses Trucks HV % HV LOS
1 11/3 299936 15:40 EB 23 74.1 31.0 186 2.4 8393 15 2504 2519 30.0% A
2 11/21 140190 7:04 SB 122 66.1 301 1806 26.1 42490 88 9113 9201 21.7% D
3 11/21 140190 7:12 NB 71 73.8 164 984 13.8 44492 68 6999 7067 15.9% B
4 11/5 269904 11:16 SB 32 66 115 460 6.5 27737 83 6434 6517 23.5% A
5 11/7 269904 12:13 NB 68 78.1 209 836 11.3 18724 48 5111 5159 27.6% B
6 11/4 730292 12:55 NB 52 78.4 132 792 10.9 no data available A
7 11/4/ 730292 13:21 NB 49 76.8 119 714 9.9 no data available A
8 11/7 970428 14:48 SB 61 71.8 113 678 9 20782 65 4036 4101 19.7% A
9 11/7/ 970428 15:20 NB 89 84.5 164 984 12.5 23811 59 3000 3059 12.8% B






37


intended to correspond to LOS ranking levels A-F. The LOS from the pilot group was

then compared to the density-based LOS calculated in accordance to the HCM. The

survey completed by the test participants can be seen in Appendix F.















CHAPTER 4
RESULTS

The main purpose of this project was to see if driver perceptions of LOS for

varying roadway and traffic conditions could be collected using videos from a driver's

perspective. Another desired result was to find out what improvements could be made to

make this study more accurate. This chapter consists of four sections. These sections

are:

* pilot test participant feedback on video clips
* pilot test participant feedback on survey instrument
* factors that were important in evaluating trip quality
* a comparison of the HCM's LOS evaluation to the LOS rankings from the pilot test

Pilot Test Participant Feedback on Video Clips

The last question on the survey given to the participants of the pilot test asked how

they would rate this exercise in terms of its ability to give them a feel for the traffic and

roadway conditions they would experience if they were actually traveling. The choices

available were excellent, very good, good, fair, poor, and very poor. Out of the seven

participants, three ranked this method very good, another three ranked this method good,

and one participant ranked this method as fair. Some suggestions and recommendations

given by the participants were to add sound, improve the video quality, videotape

different times of the day and varying weather conditions, and videotape more during

peak hour traffic conditions. Two complaints about this method were that it was hard to

determine pavement quality and it was hard to compare two-lane rural freeways to three-

lane rural freeways.









Pilot Test Participant Feedback on Survey Instrument

One of the main objectives of the pilot test was to identify the general suitability of

the survey and any potential improvements that could be made to improve the survey.

The participants of the pilot test did not offer any comments on how to improve the

survey. The survey provided the researchers with factors that the participants felt

affected their trip quality which was the intent of the survey.

Factors That Were Important in Evaluating Quality of Service

After the participant evaluated each video clip for the quality of the trip (i.e.

excellent, very good, etc.) they were asked to explain what factors led them to give that

ranking. Since there were seven participants and nine video clips there were a total of 63

surveys (7 participants x 9 videos) collected during the pilot test. The factor that was

mentioned the most in affecting the trip quality was traffic volume. Thirty-one percent of

the comments dealt with traffic volume. Pavement quality was mentioned the second

most, over twelve percent of the time. The factors that influenced the participant's trip

quality are listed below in Table 4-1, along with the percentage of time it was mentioned.

Table 4-1: Pilot Test Participants Factors Affecting Trip Quality

% of Time
Factors Influencing Trip Quality Menione
Mentioned
Traffic Volume 31.0%
Pavement Quality 12.6%
Ability to Maintain a Constant Speed 9.2%
Percentage of Trucks 9.2%
Ability to Pass/Change Lanes 8.0%
Weather Conditions 7.5%
Number of Lanes 4.6%
Other Vehicles Speeding By 4.0%
Drive at Speed Limit 2.9%
Straight Roadway (Easy to See) 2.9%
Drive Above Speed Limit 2.9%
Cars Following Closely Behind 1.7%









Lane Width 1.1%
On/Off Ramps 1.1%
Shoulders 1.1%
Scenic 0.6%

A Comparison of the HCM's LOS Evaluation to the LOS Ratings From the Pilot
Test

The LOS rankings from the pilot test were calculated by two methods. The first method

was the average response. Whichever category had the most responses was the average

survey ranking. If the majority of the seven participants rated the video clip excellent it

received a LOS rating of A. If the majority of the seven participants rated the video clip

good it received a LOS rating of C. The second method was the modal response. This

method was calculated by converting the ranking to a numerical scale first, then

averaging the rankings, and finally converting the average ranking back to a LOS letter.

The participants' results were then compared to the LOS calculated by the HCM. The

range of the participants' responses for trip quality was also noted. Out of the nine video

clips, three were rated the same as the HCM's density-based LOS. Five other clips were

rated one letter grade lower than the HCM's density-based LOS ranking. The last video

clip was rated three letter grades lower than the HCM's density-based LOS ranking.

Table 4-2 shows the comparison of the HCM's LOS evaluation to the LOS rankings from

the pilot test.

There was a correlation between some of the factors the researchers were capturing

on the video clips and what some of the participants indicated as factors influencing their

trip quality. All of the participants mentioned the number of lanes as a factor influencing

their trip quality for the video clips containing the three-lane rural freeway. For the video

clips that had a HCM density-based LOS of A participants noted the low traffic volume









as a factor influencing their decision. For the video clip that had a HCM density-based

LOS of D the participants noted high traffic volume as a factor influencing their decision.

The participants also noted percentage of trucks as a factor influencing their decision on

the videotape with the highest percentage of heavy vehicles. However, participants did

not note percentage of trucks as a factor influencing their decision when there were one

or no trucks in the video clip. Pilot test participants did not consider the rolling terrain on

the Turnpike video clips as one of the factors influencing their decision. This could have

been because the rolling terrain was hard to pick-up in the video clips.

Generally, what was seen in the video clips correlates well to the five-minute loop

data collected from the telemetered loop detectors. However, one of the segments, clip

number six, appeared to have a more significant amount of traffic at the time it was

filmed than the five-minute loop data indicated. In this case it could have been that the

five-minute interval was not small enough. Most of the data collected for the five-minute

interval could have occurred in the minute the segment was videotaped. As the other

eight video clips have shown, they accurately represent the traffic conditions that are

occurring on the rural freeway as indicated by the five-minute loop data.













Table 4-2: Comparison of Results


Modal
Avg. Response Range of
Clip Density 24 Hr HCM Survey Survey Survey
No. Date Site Time Direction (pc/hr/In) % HV LOS Rankings Rankings Rankings
1 11/3/2003 299936 15:40 EB 2.4 30.0% A B A/B A/B
2 11/21/2003 140190 7:04 SB 26.1 21.7% D D/E D/E C/E
3 11/21/2003 140190 7:12 NB 13.8 15.9% B C B/C B/C
4 11/5/2003 269904 11:16 SB 6.5 23.5% A A A/B A/C
5 11/7/2003 269904 12:13 NB 11.3 27.6% B B B/C A/D
No
6 11/4/2003 730292 12:55 NB 10.9 data A D/E C/D C/D
No
7 11/4/2003 730292 13:21 NB 9.9 data A A A/B A/B
8 11/7/2003 970428 14:48 SB 9 19.7% A B B A/C
9 11/7/2003 970428 15:20 NB 12.5 12.8% B C B/C A/C














CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS

The purpose of the pilot test was to see if a method had been developed by which

driver perceptions of LOS for varying roadway conditions could be collected. The

development of this method consisted of accomplishing six parts: vehicle

instrumentation, determination of data collection sites, collecting video data, composition

of video clips, data collection, and a pilot test. The researchers wanted to identify if the

participants were able to relate to the videos from a driver's perspective. The researchers

also wanted to find out if this was a realistic method and what improvements could be

made to make this study more accurate.

Conclusions

Vehicle Instrumentation

The instrumentation of the vehicle produced the video clips the researchers were

hoping for. All three views were captured simultaneously with good quality. The video

clips could have been more consistent if the video cameras were permanently mounted in

the test vehicle. Having to set up the test vehicle everyday not only took time but it led to

inconsistencies in the video cameras' FOV.

Determination of Data Collection Sites

The sites chosen worked well for the purpose of this study. These sites were

chosen based on their AADT and percentage of heavy vehicle data from the FTI CD.

The researchers were able to capture both two and three-lane rural freeways, low density

and medium density conditions, both low and high percentages of heavy vehicles, and









flat and rolling terrain. The researchers were unable to capture the higher range of

density conditions.

Collecting Video Data

For this study, more than one site location was videotaped in a day. This led to the

inability to capture all the varying traffic conditions at one site. Some sites were filmed

in the middle of the day when there wasn't much traffic while other sites were filmed

only during peak hour traffic conditions. The best approach would be to devote at least

one day for each site that is going to be videotaped and capture the full range of density

conditions for that particular site.

Composition of Video Clips

The video editing software program, Adobe Premier, was used to composite the

video clips. This software was able to merge three videotapes together in the layout that

the researchers desired. There were two problems with the composition of the video

clips. The main problem was when the videotapes were transformed into digital format

the digital output would sometimes freeze. This led to the second problem which was

synchronization between the three camera views. If one of the views froze it would

become off sync from the other two views.

Data Collection

The traffic information was provided from the telemetered loop detectors in five-

minute intervals. Generally, what was seen in the video clips correlated well to the five-

minute loop data collected from the telemetered loop detectors. The alternative to using

the telemetered loop data is to determine if there is enough correlation with what can be

seen through the video to use what can be seen as the traffic data.









Pilot Test

From the pilot test most of the participants were able to relate to the videos as if

they were the driver. There were a few situations where participants commented that

they would have driven differently, but for the most part the participants said they were

able to relate to being the driver. Most of the participants also felt that this was a realistic

method for what the researchers were trying to achieve.

The results of the pilot test showed that subjects' LOS rankings were not far from

the density-based HCM's LOS ranking. Out of the seven participants, no one used

density as the only criterion when rating LOS. All participants considered more than just

density.

Recommendations

The method of showing subjects videos as if they were the driver was successful.

The overall video quality used in the pilot test could be improved. Also, another way to

better show pavement quality on the videos should be investigated. Since density was

not the only consideration by the participants in this study, it is recommended that a more

detailed study using this research approach be conducted. It is hoped that a more detailed

study will provide the means to develop a procedure that will lead to specific

recommendations for rural freeway LOS criteria and thresholds.














APPENDIX A
COMMON ABBREVIATIONS

HCM Highway Capacity Manual

LOS Level of Service

TRB Transportation Research Board

HCQS Highway Capacity and Quality of Service

QOS Quality of Service

LF Load Factor

v/c Volume to Capacity Ratio

AID Average Individual Delay

WDOT Wisconsin Department of Transportation

MOE Measures of Effectiveness

pc/mi/ln Passenger Cars per Mile per Lane

NA Not Applicable

FOV Field of View

FDOT Florida Department of Transportation

ILD Inductance Loop Detectors

FIHS Florida Intrastate Highway System

AADT Average Annual Daily Traffic

GIS Geographic Information System

FTI Florida Traffic Information















APPENDIX B
POSSIBLE SITE LOCATION TRAFFIC INFORMATION












Type


Site


"K" "D" "T"
Factor Factor Factor


30191

30351

100106

100110

100123

100194

100224

00 109922

109926

120184

140190

150183

170225

189920

269904

290269

290320


Description
SR-93/1-75,0.5 MI N OF CR-896,COLLIER CO.

SR-93/1-75,W OF EVERGLADES BLVD,COLLIER CO.
SR-400/1-4, UNDER BETHLEHAM RD OVERPASS,
HILL CO

1-275, 1.3 MI E OF HOWARD FRANKLIN BR
1-275 AT S. END OF FLORIBRASKA AVE OVERP-
BAD LP
1-75, 0.6 M S OF US301, 2.3 M N OF 1-4, HILLS--BAD
LOOP
1-75, 0.9 MI S OF SR60 AT SR618 O.P.,
HILLSBOROUGH CO
SR-93/I-275, 0.25 MI N OF FLETCHER AV, TAMPA,
HILL CO
1-75, 1.25 MI N OF SR-60 (ADAMO DR.), TAMPA,
WIM#26
SR-93/I-75, 225' S OF DANIELS PKWY UNDERPASS,
LEE
1-75, 0.6 MI. SOUTH OF SR-54, PASCO CO. -
UC10/28/94
1-275, 300 YDS S OF THE SB TOLL, PINELLAS-
UC11/93
1-75, @ PROCTOR RD OP, 0.7 MI N SR 72,
SARASOTA CO

1-75, 3.5 MI S OF FLORIDA TURNPIKE, WIM#20
1-75/SR-93, 3 MILES NORTH OF MARION COUNTY
LINE

1-10, 0.45 MI EAST OF US41, LAKE CITY
1-75 NORTH OF LAKE CITY BETWEEN 1-10 AND US-
90


AADT
Direction Direction Two-
1 2 Way
27,043 54,616
N 27,573 S C
N C
9,169 E 9,182 W 18351
C
48,583 50,946 99,529
E W C
62,329 62,426 124,755
E W C
65,831
C
49,231 97,891
N 48,660 S C
51,358 103,845
N 52,487 S C
66,000
S
57,374 111,743
N 54,369 S C
24,674 46,667
N 21,993S C
35,427 70,702
N 35,275 S C
22,171 45,019
N 22,848 S C
42,732 85,595
N 42,863 S C
39,000
F
29,922 58,687
N 28,765 S C
10,164 10,028 20,192
E W C
22,340 45,131
N 22,791 S C


9.88 A

12.94 A

8.30 A

8.46 A

9.20 D

9.10D

9.20 D

9.20 D

9.20 D

9.73 A

8.99 A

9.57 A

9.83 A

10.97 D

12.50 D

12.27 A

13.62 A


53.63 A

56.13 A

54.20 A

55.20 A

53.92 D

53.50 D

53.92 D

53.92 D

53.92 D

57.77 A

56.15 A

52.30 A

53.85 A

55.06 D

57.01 D

67.04 A

54.44 A


13.05 D

11.46 A

19.36 D

8.71 D

8.71 D

9.35 A

9.82 A

8.71 D

7.05 A

13.05 D

19.36 D

5.87 A

18.10 D

17.99 D

21.85 A

25.89 A

25.89 A











Site
Site Type

299936 T

320112 T

360317 T

370238 T

370352 T

480156 T

480560 T

489924 T

500220 T

530218 T
549901 T

550208 T

550304 T

570318 T

609928 T

610152 T


Description
1-10, 50 FT WEST OF CR-250 OVERPASS, LAKE
CITY

1-75, S OF STATE LINE NORTH OF S.R. 143
1-75, SB SHOULDER, 0.35 MILES N OF WILLIAMS
RD.

1-10, 0.15 MILES WEST OF CR 136
1-10, SUWANNEE CO, 1.5 MI WOF ELLAVILLE
SCALES

1-10, 1.5 MI WEST OF US-90

SR-8/I-10, 1 MILE OF SR-291/DAVIS HWY (RTMS)
1-110, 1 MI S OF 1-10, PENSACOLA, WIM#24--UC
10/94
1-10, 250 FT W OF CR-268 OVERPASS, GADSDEN
CO.
1-10, 1 MI. EAST OF US-231, JACKSON CO. -- UC
11/94
SR-8/I-10,0.66 MI E OF CR-257,JEFFERSON CO.
MISSION RD,NORTH OF I-10,TALLAHASSEE,LEON
CO.
SR-8/I-10,1 MI WOF THOMASVILLE RD U/P,LEON
CO.
SR-8/I-10,@ANTIOCH RD O/P,OKALOOSA CO.
SR-8/I-10,1.3 MI WEST OF BOY SCOUT RD,WALTON
CO.
SR-8/I-10,AT CR-273,SE OF CHIPLEY,WASHINGTON
CO.


AADT
Direction Direction Two-
1 2 Way
10,104 10,183 20,287
E W C
18,515 37,198
N 18.683 S C
37,579 74,516
N 26,937 S C
12,725 12,621 25,346
E W C

8,124 C
16,073 14,564 30,637
E W C
20,538
OE W OC
26,390 52,441
N 26,051 S C
13,664 13,337 27,001
E W C
10,821 10,768 21,589
E W C
12,478 12,527 25,005
E W C
4,896 N 4,916S 9,812 C
28,625 28,275 56,900
E W C
10,347 10,504 20,851
E W C
9,590 E 9,526 W 19116
C
9,320 E 9,355 W 18,675
C


17,780


700134 T SR-9/I-95,3.34 MI. S. OF SR-514,BREVARD CO.


"K"
Factor

12.50 D

15.17 A

11.69 A

11.89 A

12.50 D

10.56 A

9.64 D

9.64 D

11.68 A

12.90 A
13.69 A

11.18D

9.64 A

10.65 A

12.05 D

12.41 A


"D"
Factor

57.01 D

54.98 A

54.81 A

53.84 A

57.01 D

58.54 A

51.15D

51.15D

54.72 A

52.53 A
52.46 A

57.23 D

51.15 A

52.21 A

55.16 D

58.45 A


"T"
Factor


27.12 A

29.93 A

17.99 D

22.64 A

29.61 A

26.51 D

14.60 D

4.05 A

18.22 A

26.51 D
22.78 A

5.3 D

12.44 A

19.68 A

21.33 A

21.51 A


18,070 S 35,850 10.97 D 55.06 D 14.88 A











Site
Site Type
700322 T

709919 T

729923 T

720354 T

720109 T

720171 T

729914 T

729905 T

730292 T

740132 T

770343 T

790133 T

799906 T

860163 T

860186 T

860357 T

870108 T


Description
SR-9/1-95,0.9 MI S OF AURANTIA RD U/P,BREVARD
CO.
SR-9/1-95,3.45 MI S OF SR-514,MALABAR,BREVARD
CO.
SR-9/1-95,0.75 MI S OF DUNN AVE,JAX,DUVAL CO.

SR-8/1-10,1 MI E OF MCDUFF AV,JAX,DUVAL CO.
SR-8/I-10,@CR-217 OVERPASS,E. OF
BALDWIN,DUVAL CO.
SR-9/1-95,0.7 MI N OF UNIVERSITY
BLVD,JAX,DUVAL CO
SR-9A/I-295,3 MI N OF I-10,JACKSONVILLE,DUVAL
CO.
SR-9/1-95,2 MI S OF 1-295 S INTERCHANGE,DUVAL
CO.
SR-9/1-95,1.4 MI S OF PALM COAST
PKWY,FLAGLER CO.
SR-9/1-95,2.0 MI S OF GA. STATE LINE,NASSAU CO.

SR-400/1-4,1.6 MI E OF SR-434,SEMINOLE CO.

1-95,2.7 MI N OF SR44,@CR44 O/P,VOLUSIA CO.
ON 1-4,169' E OF ENTERPRISE RD O/P,VOLUSIA
CO.
SR-9/I-95,@NE 48TH ST,POMPANO BCH,BROWARD
CO.
SR-862/I-595,0.2 MI E OF UNIVERSITY
DR,BROWARD CO.
SR-93/I-75,2 MI WOF US-27,.6 MI W
TOLL,BROWARD CO
SR-112/I-195,1600' E OF SR-5/US-1,DADE CO.


AADT
Direction Direction Two-
1 2 Way
14,566 1601 29,167
14,601 S
N C
17,994 1726 35,720
17,726 S
N C
69,500
S
68,110 69,028 137,138
E W C
22,665 22,429 45,094
E W C
55,386 110,767
N 55,381 S C
N C
31,128 3 61,791
N 30,663 S C
N C
34,405 33453 67,858
33,453 S
N C
28,115 2 56,736
N 28,621 S C
N C
26,031 76 51,792
N 25,761 S C
N C
60,981 60,669 121,650
E W C
20,046 1745 39,791
N 19,745 S
N C
0E W 76,000
F
99,069 9696 198,765
99,696 S
N C
91,065 86,047 177,112
E W C
10,043 10,026 20,069
E W C
48,348 47,387 95,735
E W C


"K"
Factor
11.32 A

10.97 D

9.06 D

8.83 A

9.54 A

9.11 A

9.24 A

9.06 D

10.14 A

12.5 D

8.06 A

9.89 A

8.61 D

7.83 A

8.32 A

12.22 A

8.47 A


"D"
Factor
54.06 A

55.06 D

55.55 D

61.55 A

54.73 A

52.57 A

52.52 A

55.55 D

55.85 A

57.01 D

52.46 A

56.31 A

54.73 D

51.22 A

55.82 A

58.41 A

52.54 A


"T"
Factor

18.7 A

16.84 A

10.57 D

4.15A

22.13 A

5.77 A

14.66 A

10.57 D (
o
8.94 D

26.82 D

6.41 A

13.35 A

8.94 D

6.83 A

4.7 A

13.42 A

2.35 D











Site
Site Type Description
ON 1-4,0.5 MI SWOF ORANGE CO LINE,OSCEOLA
920303 T
CO.
930198 T SR-9/I-95,@SW23RD AVE O/P,1.5 MI S SR-
804,PALM BCH
SR-9/1-95,0.8 MI N OF DONALD ROSS RD,PALM
930217 T
BEACH CO
940260 T SR-9/1-95,0.6 MI S OF SR-68/ORANGE AV,ST LUCIE
CO.
940334 T SR-9/1-95,@MARTIN CO LINE,ST. LUCIE CO.


Direction Direction
1 2
54,281 55,142
E W


AADT
Two-
Way
109,423
C
159,000
S


41,368 ,333 83,701
42,333 S
N C
42,500
F
21,475 43,823
N C


"K"
Factor
7.63 A


"D"
Factor
55.88 A


"T"
Factor

7.42 A


8.56 D 55.24 D 10.8 D

8.81 A 61.12 A 10.17 A

8.56 D 55.24 D 10.8 D

9.28 A 52.79 A 19.94 A


Site Type: T = Telemetered; P = Portable
AADT Flags: C = Computed; E = Manual Estimate; F = First Year Est; S = Second Year Est; T = Third Year Est; X = Unknown
"K/D" Flags: A = Actual; F = Volume Fctr Catg; D = Dist/Functional Class; S = State-wide Default; W = One-Way Road
"T" Flags: A = Actual; F = Axle Fctr Catg; D = Dist/Functional Class; S = State-wide Default; X = Cross-
Reference















APPENDIX C
FINAL SITE LOCATION DETAILS














-in SS Sw
FKM SO W

AADT TruckAADT
TRAFFIC-AV TRuCKAT
0- 15,O0 0 -0 4
~0 11 7011 111 4
7 01 1 ... 1000 !i- 1gg g
or 130.001 -6000.tUS



i i Created 10/2003

29P269 '99936-. FDOT
j ~293
29coS32fl- ~. 1 DisriivL 2


\ {
'A7"













Site Loation


\\/


Ii










V


730?92


S AADT Values


r


* 1l',4rI 4 rtio

AADT Truck AADT
TRAFFIC AV TRUCKAADT
- 0- 15.00 40599
- MAI,00.a00m ----. 01499
-- .4I 1- O .---..-... 5002999
7.01:30,R 0'.... 3000-5999
- 131.00 600-PLUS
N


I Created 1012003
S FDOT
\ District 5

\i, -'.








S-i- \ -





Truk 4DT64 ies


j


"",
















Site Location 5r



AADT Truck AADT
TRAFICAV TRUCKAADT
S0 490
15.01, ge a w 5000 1499
-- rs01 n oo ------ 2990




i- i'~~ -- ,,.1. ,, ICAPT0oss
Created 1012003

-- FDOT
269904 A *'Di3Ljcl 2


















St9 FlsceUaionrm

AADT Truck AADT
TRAFFIC AV TRUCKAADT
0-150 049
5.0a -------- 1400 299
g 0 \aa o "0-000 3000-599
\ M s000 -PLUS




S\ Created 10/2003
:- A -,-'-, FDOT



II I 0-\
I II


'i D















Loc Faitionecar7
, / / / .......e.hn ty.
AADT
TRAFFIC A Truck AADT
S_ TRUCKAADT

-....00--' 3000
u s 0000 PLU &



f ) N' --A -4-.--
Created 1012003

~- FDOT
/;1 V\' 'D- 1n T- District 5
970428 I

-




















AADT
flRAFFIC AV Truck AADT
TRUCKAADT
01.0 14 90
3 8601 70,000 500 1499
00I,130 003000 -099
110,401%3.3 00 20
0000 -PLU&




--Created 1012003


~v District 7






i I' i'



-i /



OF I
-1.40 490 Values
L-




ATVle T.rcATVle















APPENDIX D
SITE LOCATIONS AND DRIVING TIMES










# Site # of Lanes Peak Hour Peak Hour
Route Number Direction N, E Direction S, W Direction MP Date Time Direction
1 1-10 290269 2 6 p.m. 5 p.m. 302-301 11/3/2003 3:30 p.m. EB
4:09 p.m. WB
4:32 p.m. EB
5:08 p.m. WB
5:40 p.m. EB
2 1-10 299936 2 6 p.m. 5 p.m. 312-311 11/3/2003 3:40 p.m. EB
4:02 p.m. WB
4:39 p.m. EB
5:01 p.m. WB
5:48 p.m. EB
6:09 p.m. WB
3 1-75 290320 3 4 p.m. 5 p.m. 428-429 11/3/2003 3:24 p.m. NB
4:18 p.m. SB
4:24 p.m. NB
5:17 p.m. SB
5:33 p.m. NB
4 1-95 730292 2 5 p.m. 4 p.m. 288-289 11/4/2003 12:55 p.m. NB
1:13 p.m. SB
1:21 p.m. NB
1:38 p.m. SB
1:46 p.m. NB
4:56 p.m. SB
5:04 p.m. NB
5:22 p.m. SB
5:35 p.m. NB
5 1-75 269904 3 1 p.m. & 4 p.m. 4 p.m. 376 11/5/2003 11:01 a.m. SB
11:16 a.m. SB
11:20 a.m. NB









Site # of Lanes Peak Hour Peak Hour Direction
# Route Number Direction N, E Direction S, W Direction MP Date Time
11:05 a.m. NB
11:32 a.m. SB
11/7/2003 11:55 a.m. NB
12:09 p.m. SB
12:13 p.m. NB
12:28 p.m. SB
4:48 p.m. NB
6 1-75 189920 2 325-324 11/5/2003 12:13 p.m. SB
12:19 p.m. NB
12:29 p.m. SB
11/7/2003 1:09 .m. SB
1:16 p.m. NB
1:28 p.m. SB
1:34 p.m. NB
11/9/2003 3:35 p.m. NB
7 Turnpike 970428 2 6 p.m. 6 p.m. 281 11/5/2003 3:21 p.m. SB
4:18 p.m. NB
4:33 p.m. SB
4:50 p.m. NB
11/7/2003 2:48 p.m. SB
3:20 p.m. NB
3:28 p.m. SB
8 1-75 140190 2 6 p.m. 8 a.m. 276-275 11/5/2003 1:58 p.m. NB
2:11 p.m. SB
2:17 p.m. NB
11/9/2003 3:00 p.m. NB
11/21/2003 6:37 a.m. NB
6:41 a.m. SB










Site # of Lanes Peak Hour Peak Hour Direction
# Route Number Direction N, E Direction S, W Direction MP Date Time
6:48 a.m. NB
6:52 a.m. SB
6:59 a.m. NB
7:04 a.m. SB
7:12 a.m. NB
7:15 a.m. SB
2:11 p.m. SB
2:18 p.m. NB
2:22 p.m. SB
2:28 p.m. NB
2:32 p.m. SB
2:39 p.m. NB















APPENDIX E
LOOP DATA SUMMARY TABLES










Date Total Avg. 5 min
Tag Site Time Direction Volume Speed vol veh/hr/ln density

11/3/2003 SPD 290320 (1-75, near 1-10) NO DATA AVAILABLE
11/3/2003 SPD 290269 (I-10, near 1-75 NO DATA AVAILABLE
11/3/2003 SPD 299936 (I-10, near 1-75) 15:40 Eastbound 23 74.1 31.0 186 2.4
16:39 27 74.3 31 186 2.5
17:48 24 71 32 192 2.6
16:02 Westbound 15 76.3 55 330 4.4
17:01 11 78 44 264 3.5
18:09 4 75.5 25 150 2.1

11/4/2003 SPD 730292 (1-95, near Daytona) 12:55 Northbound 52 78.4 132 792 10.9
13:21 49 76.8 119 714 9.9
13:46 39 78.3 120 720 9.8
17:04 36 78.2 106 636 8.5
17:35 56 78 141 846 11.6
13:13 Southbound 77 71.5 116 696 9.6
13:38 73 70.6 135 810 11
16:56 75 72.7 147 882 11.9
17:22 91 71.9 147 882 11.7


11/5/2003 SPD 269904 (I-75, near Micanopy) 11:05 Northbound 19 80.9 116 464 6.3
11:20 30 80 136 544 7.3
11:01 Southbound 41 66 144 576 8.4
11:16 32 66 115 460 6.5
11:32 44 65.6 148 592 8.4










Total Avg. 5 min density
Date Tag Site Time Direction Volume Speed vol veh/hr/ln
11/5/2003 SPD 189920 (1-75, s. of Turnpike merge) 12:19 Northbound NO DA-TA .\\-\AILA-BLE
12:13 Southbound NO DATA AVAILABLE
12:29 NO DATA .\\-\AILABLE

11/5/2003 SPD 140190 (1-75, near Tampa) 13:58 Northbound 75 73.37 137 822 11.6
14:17 79 74.01 164 984 14
14:11 Southbound 92 67.08 160 960 13.5

11/5/2003 SPD 970428 (Turnpike, near Orlando) 16:18 Northbound 54 78.4 107 642 8.8
16:50 54 79.6 105 630 8.4
15:21 Southbound 45 69.8 73 438 6
16:33 55 69.5 112 672 9.3

11/7/2003 SPD 269904 (1-75, near Micanopy) 11:55 Northbound 48 79.2 180 720 9.6
12:13 68 78.1 209 836 11.3
16:48 120 79.1 293 1172 15.5
12:09 Southbound 48 69 195 780 10.8
12:48 39 67.1 163 652 8.9

11/7/2003 SPD 189920 (1-75, s. of Turnpike merge) 13:16 Northbound D.-TA AT ONE HOUR INTER\AL
13:34 DATA AT ONE HOUR INTERNAL
13:09 Southbound DATA AT ONE HOUR INTER\VAL
13:28 DATA AT ONE HOUR INTER\XAL

11/7/2003 SPD 970428 (Turnpike, near Orlando) 15:20 Northbound 89 84.5 164 984 12.5
14:48 Southbound 61 71.8 113 678 9










Total Avg. 5 min density
Date Tag Site Time Direction Volume Speed vol veh/hr/ln
15:28 60 72.1 123 738 9.8
11/9/2003 SPD 140190 (1-75, near Tampa) 15:00 Northbound NO DATA .\\-\IL-\BLE


11/9/2003 SPD 189920 (I-75, s. of Turnpike merge) 15:35 Northbound DATA AT ONE HOUR INTERVAL


11/21/2003 SPD 140190 (I-75, near Tampa) 6:37 Northbound 54 75.5 135 810 11.2
6:48 67 76.4 166 996 13.8
6:59 56 74.4 145 870 12.2
7:12 71 73.8 164 984 13.8
14:18 102 73.5 236 1416 20.2
14:28 158 71.2 274 1644 24.1
14:39 101 74 228 1368 19.2
6:41 Southbound 123 65.3 306 1836 26.7
6:52 124 65.1 297 1782 26
7:04 122 66.1 301 1806 26.1
7:15 135 65.5 302 1812 26.3
14:11 88 72 211 1266 17.1
14:22 96 70.8 214 1284 17.4
14:32 89 67.1 227 1362 19.2










Total 24 Hr
5 min veh/hr/ln 24 Hr 24 Hr 24 Hr 24 Hr %HV
Date Tag Site Time Direction vol HV HV Vol Buses Trucks HV

11/3/2003 CLS 290320 (1-75, near I-10) i a bc
no data n a\nllablc'
11/3/2003 CLS 290269 (I-10, near 1-75)
11/3/2003 CLS 299936 (I-10, near 1-75) 15:40 Eastbound 4 24 Total 17277 31 5057 5088 29.4%
16:39 6 36
17:48 12 72 WB 8884 16 2553 2559 28.8%
16:02 Westbound 21 126
17:01 12 72 EB 8393 15 2504 2519 30.0%
18:09 7 42
730292 (1-95, near
11/4/2003 CLS Daytona) 12:55 Northbound
13:21
13:46
17:04
17:35 o data a\ ailablc
13:13 Southbound
13:38
16:56
17:22
269904 (1-75, near
11/5/2003 CLS Micanopy) 11:05 Northbound 25 150 Total 50443 104 13439 13543 26.8%
11:20 34 204
11:01 Southbound 32 192 NB 19320 40 6699 6739 34.9%
11:16 22 132
11:32 34 204 SB 31123 64 6740 6804 21.9%










Total 24 Hr
5 min veh/hr/ln 24 Hr 24 Hr 24 Hr 24 Hr %HV
Date Tag Site Time Direction vol HV HV Vol Buses Trucks HV
189920 (1-75, s. of
11/5/2003 CLS Turnpike merge) 12:19 Northbound T 35t2 7l 1|5i4 1I5,2 7"..
12:13 Southbound data at one h0ou NB lIu5s 3' 533 5423 32 '"..
iitei' al
12:29
_SB 1s724 4-I 5111 515u 27 i".
140190 (1-75, near
11/5/2003 CLS Tampa) 13:58 Northbound 29 174 Total 72280 78 12171 12249 16.9%
14:17 23 138
14:11 Southbound 30 180 NB 34775 36 6139 6175 17.8%


SB 37505 42 6032 6074 16.2%
970428 (Turnpike, near
11/5/2003 CLS Orlando) 16:18 Northbound 11 66 Total 29694 81 7823 7904 26.6%
16:50 30 180
15:21 Southbound 16 96 NB 43 2748 2791
16:33 16 96
269904 (1-75, near
11/7/2003 CLS Micanopy) 11:55 Northbound 31 186 Total 76912 204 12112 12316 16.0%
12:13 32 192
16:48 30 180 NB 27737 83 6434 6517 23.5%
12:09 Southbound 21 126
12:48 25 150 SB 49175 121 5678 5799 11.8%










Total 24 Hr
5 min veh/hr/ln 24 Hr 24 Hr 24 Hr 24 Hr %HV
Date Tag Site Time Direction vol HV HV Vol Buses Trucks HV
189920 (I-75, s. of To
11/7/2003 CLS Turnpike merge) 13:16 Northbound da a 51u- I4-> 14 NIX 11 11111 II 3".
data at one Ihour
13:34 intcerial
13:09 Southbound NB 2744 1-' i 5315 53'4 IL jI"..
13:28
SB 244-u7 77 4547 41-24 IN "'..
970428 (Turnpike,
11/7/2003 CLS near Orlando) 15:20 Northbound 14 84 Total 44593 124 7036 7160 16.1%
14:48 Southbound 11 66
15:28 19 114 NB 23811 59 3000 3059 12.8%


SB 20782 65 4036 4101 19.7%
140190 (1-75, near
11/9/2003 CLS Tampa) 15:00 Northbound
_no data an ailable




189920 (I-75, s. of
11/9/2003 CLS Turnpike merge) 15:35 Northbound Total 45lu, 12L 44\4 4-12 1, ,,"..
Sdata at one hour _
intcrlial NB I IlI s II u I|l7X 1737 I 1".,


SB 2'I> 2 5u 2?l1 \ 2 75 I1i 7".,










Total 24 24 Hr
5 min veh/hr/ln 24 Hr Hr 24 Hr 24 Hr %HV
Date Ta Site Time Direction vol HV HV Vol Buses Trucks HV
140190 (1-75, near
11/21/2003 CLS Tampa) 6:37 Northbound 23 138 Total 86982 156 16112 16268 18.7%
6:48 33 198
6:59 36 216 NB 44492 68 6999 7067 15.9%
7:12 30 180
14:18 31 186 SB 42490 88 9113 9201 21.7%
14:28 42 252
14:39 35 210
6:41 Southbound 98 588
6:52 109 654
7:04 97 582
7:15 104 624
14:11 24 144
14:22 30 180
14:32 28 168















APPENDIX F
PILOT TEST SURVEY








UNIVERSITY OF
SFLORIDA


TriC


Transportation Research Center


About Yourself



Gender: 0 Male D Female

Age: 1 16 to 25 years I 26 to 45 years I 46 to 65 years
D Over 65 years

Highest level of education:
0 Some or no high school I High school diploma or equivalent
I Technical college degree (A.A.) D College degree
D Post-graduate degree

Approximate annual household income:
0 No income I Under $25,000 0 $25,000 49,999
1 $50,000 74,999 1 $75,000 99,999 1 $100,000 149,999
1 $150,000 or more

Number of years possessing a driver's license:


About Your Rural Freeway Drivina


Typical number of rural freeway trips made during a month?


1 to 2
Over 12


3 to 4 5 5 to 6


7 to 8 O 9 to 10


Typical length of trip made on rural freeway (in miles)?


less than 16 miles
76 to 100
Over 200


16 to 30
101 to 125


31 to 45
126 to 150


46 to 60
151 to 175


Vehicle type most often used for rural freeway trips:


Sports car
Full-size van


Pick-up truck
RV/Motorhome


Other


11 to 12


Sedan


Mini-van
motorcycle


61 to 75
176 to 200


SUV













I Your ODinions

Rank the overall quality of the trip for the conditions observed in the video clip. Place an
'X' in the appropriate column of the table below.


List all the factors/reasons that influenced your ranking of the trip quality for this video
clip. After listing the factors, please number them from most important to least important












In general, how would the purpose of your trip (such as business, recreational, social)
affect the trip quality rankings assigned above?











How would you rate this exercise in terms of its ability to give you a feel for the traffic
and roadway conditions you would experience if you were actually driving?


Very Good I Good I Fair


Poor I Very Poor


Excellent















LIST OF REFERENCES


1. Transportation Research Board. (2000). Special Report 209: Highway Capacity
Manual. Transportation Research Board, Washington, D.C., 4th Ed.

2. Harwood, D., Flannery, A., McLeod, D., Vandehey, M. (July 2001). The Case for
Retaining the Level of Service Concept in the Highway Capacity Manual.
Presented at the 2001 Transportation Research Board Committee A3A10 -
Highway Capacity and Quality of Service Midyear Meeting, Truckee, California.

3. Transportation Research Board. (1985). Special Report 209: Highway Capacity
Manual. Transportation Research Board, Washington, D.C.

4. Washburn, S. (2003). Facility Performance Model Enhancements for Multimodal
Systems Planning. Transportation Research Center at the University of Florida,
Gainesville, Florida.

5. Pecheux, K., Tarko, A., Rabbani, E., Hall F. (July 2001). Scouting Party on User
Perceptions and LOS Performance Measures. Presented at the 2001 Transportation
Research Board Committee A3A10 Highway Capacity and Quality of Service
Midyear Meeting, Truckee, California.

6. Sutaria, T.C., and Haynes, J.J. (1977). Level of Service at Signalized Intersections.
Transportation Research Record 644, Transportation Research Board, Washington,
D.C., 107-113.

7. Highway Research Board. (1965). Special Report 87: Highway Capacity Manual.
National Research Council, Washington, D.C.

8. Pecheux, K., Pietrucha, M., Jovanis, P. (2000). User Perception of Level of Service
at Signalized Intersections: Methodological Issues. Transportation Research
Circular E-C018: Proceedings of the Fourth International Symposium on Highway
Capacity, National Research Council, Washington, D.C., 322-335.

9. Hall, F., Wakefield, S., Al-Kaisy, A. (2000). Freeway Quality of Service: What
Really Matters to Drivers and Passengers? Transportation Research Record:
Journal of the Transportation Research Board, No. 1776, Transportation Research
Board, National Research Council, Washington, D.C., 17-23.

10. Hostovsky, C., Hall, F. (2003). Freeway Quality of Service: Perceptions from
Tractor-Trailer Drivers. Presented at the Transportation Research Board 82rd
Annual Meeting, Washington, D.C.









11. Hostovsky, C., Wakefield, S., Hall, F. (2003). Mitigating Traffic Congestion
Impacts: Users' Perceptions of the Quality of Transportation Service. Submitted to
the Canadian Geographer for publication.

12. Nakamura, H., Suzuki, K., Ryu, S. (2000). Analysis of the Interrelationship Among
Traffic Flow Conditions, Driving Behavior, and Degree of Driver's Satisfaction on
Rural Motorways. Transportation Research Circular E-C018: Proceedings of the
Fourth International Symposium on Highway Capacity, National Research Council,
Washington, D.C., 42-52.

13. Pfefer, R. (2000). Toward Reflecting Public Perception of Quality of Service in
Planning, Designing and Operating Highway Facilities. Transportation Research
Record: Journal of the Transportation Research Board No. 1685, Transportation
Research Board, National Research Council, Washington, D.C., 53-62.

14. Stein-Hudson, K. (1995). Customer-Based Quality in Transportation. NCHRP
Report 376, Transportation Research Board, National Research Council,
Washington, D.C.

15. Florida Traffic Information 2002. (2002). Florida Department of Transportation,
Tallahassee, FL, CD-ROM.

16. Users Guide for Adobe Premiere Pro Software. (n.d.). Retrieved November 17,
2003, from http://www.adobe.com/products/premiere

17. Transportation Research Board. (1998). Special Report 209: Highway Capacity
Manual. Transportation Research Board, Washington, D.C., 3rd Ed.

18. Kittelson, W. (2001). Highway Capacity and Quality of Service. Presented at the
2001 Transportation Research Board Committee A3A10 Highway Capacity and
Quality of Service Midyear Meeting, Truckee, California.















BIOGRAPHICAL SKETCH

I am a 23 year old graduate student at the University of Florida. I received my

undergraduate degree of Bachelor of Science in Civil Engineering from the University of

Florida in May of 2002. I am currently working on my Master of Science from the

College of Engineering, specializing in transportation.