Citation
Operational Effects to Different Transportation Modes at Signalized Intersections from Differing Geometries, Signal Systems, and Volume Levels

Material Information

Title:
Operational Effects to Different Transportation Modes at Signalized Intersections from Differing Geometries, Signal Systems, and Volume Levels
Creator:
Valila, Tyler J
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (121 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.E.)
Degree Grantor:
University of Florida
Degree Disciplines:
Civil Engineering
Civil and Coastal Engineering
Committee Chair:
ELEFTERIADOU,AGELIKI
Committee Co-Chair:
WASHBURN,SCOTT STUART
Committee Members:
SRINIVASAN,SIVARAMAKRISHNAN

Subjects

Subjects / Keywords:
bicyclist -- intersections -- operational -- pedestrian -- signal -- simulation -- transportation -- vehicle
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Civil Engineering thesis, M.E.

Notes

Abstract:
In the United States, walking and bicycling have become larger shares of transportation. Signalized intersections have not changed to a high degree geometrically or technologically to accommodate larger volumes of these modes in the presence of motor vehicles. The most advanced signal systems used today do not take pedestrians or bicyclists into account as being equal to the motor vehicle. Simple steps can be taken at the signalized intersection to improve serviceability for all users. The first objective of this research is to identify safe designs for pedestrians and bicyclists at signalized intersection. The second objective is to create and simulate these intersection characteristics with altering geometries, signal systems, and volumes to determine which combinations produce reduction in delay. The third objective is to create a set of guidelines based on the results that summarizes these operational effects. This research is meant to take an in depth look at design considerations and resulting operational effects that an engineer will need to consider for intersection betterment projects. The methodology involves building common scenarios seen in a signalized intersection using a micro simulator and finding which designs can improve intersection functionality the best. All results were yielded from the micro simulator VISSIM. Measures of effectiveness include delay, vehicle queueing, and average vehicle speed. These measurements were compared to base data to look for relative impacts. The results show good improvements are possible to an intersection with simple modifications. Implementing set back cross walks can reduce vehicle delay. Downstream crossing points for bicyclists can increase vehicle average speed. A combination of these features gives favorable results for weighted mode delay. Finally, staggered crosswalks at intersections give little to no operational benefits. This research was performed solely in simulation and should be confirmed with real world field results. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.E.)--University of Florida, 2017.
Local:
Adviser: ELEFTERIADOU,AGELIKI.
Local:
Co-adviser: WASHBURN,SCOTT STUART.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2018-06-30
Statement of Responsibility:
by Tyler J Valila.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
6/30/2018
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 OPERATIONAL EFFECTS TO DIFFERENT TRANSPORTATION MODES AT SIGNALIZED INTERSECTIONS FROM DIFFERING GEOMETRIES, SIGNAL SYSTEMS, AND VOLUME LEVELS By TYLER J VALILA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2017

PAGE 2

2 2017 Tyler J. Valila

PAGE 3

3 To my parents, Paul and Victoria Valila

PAGE 4

4 ACKNOWLEDGEMENTS I have many people to thank for their help and support in completing graduate school and this thesis. First, I would like to thank my friends and family, especially my parents, for their unwavering support throughout my journey in school. I deeply thank the friends who gave me advice on how to deal with such a large workload and guidance in general. I thank Nithin Agarwal Pruth vi Manjunatha and Ryan Casburn for their help teaching me how to use VISSIM. I thank the professors at the University of Florida Transportation Institute, especially my thesis committee members, Dr. Scott Washburn and Dr. Siva Srinivasan. Most of all, I would like to thank my supervisory committee chair and advisor, Dr. Lily Elefteriadou, for her support, guidance, and help throughout the course of this study.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURE S ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Background ................................ ................................ ................................ ............. 15 Research Objectives ................................ ................................ ............................... 16 Document Organization ................................ ................................ .......................... 16 2 LITERATURE REVIEW ................................ ................................ .......................... 17 Signal Control and Phasing ................................ ................................ .................... 17 Ve hicles ................................ ................................ ................................ ............ 17 Bicyclists ................................ ................................ ................................ ........... 21 Pedestrians ................................ ................................ ................................ ...... 24 Detection ................................ ................................ ................................ ................. 26 Vehicles ................................ ................................ ................................ ............ 27 Bicyclists ................................ ................................ ................................ ........... 28 Pedestrians ................................ ................................ ................................ ...... 29 Intersection Design ................................ ................................ ........................... 31 Summary ................................ ................................ ................................ ................ 34 3 METHODOLOGY ................................ ................................ ................................ ... 35 Intersection Configuration ................................ ................................ ....................... 35 Simulation ................................ ................................ ................................ ............... 41 Scenarios ................................ ................................ ................................ ................ 42 Performance Measures ................................ ................................ ........................... 46 Statisti cal Comparison Process ................................ ................................ .............. 46 4 RESULTS ................................ ................................ ................................ ............... 49 Delays ................................ ................................ ................................ ..................... 49 Vehicle Queues ................................ ................................ ................................ ...... 57 Average Vehicle Speed ................................ ................................ .......................... 63

PAGE 6

6 5 CONCLUSIONS AND RECOMMENDATIONS ................................ ....................... 69 APPENDIX: RESULT SHEETS ................................ ................................ ..................... 74 LIST OF REFERENCES ................................ ................................ ............................. 119 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 121

PAGE 7

7 LIST OF TABLES Table page 3 1 Criteria table for VISSIM analysis. ................................ ................................ ...... 45 3 2 Volume table for vehicles (vehicles per hour). ................................ .................... 46 3 3 Volume table for pedestrians and bicyclists (individuals per hour). .................... 46 3 4 Criteria table for comparative analysis. ................................ ............................... 48 5 1 Summarized results by configuration. ................................ ................................ 70 A 1 Base, pre timed, medium veh, low bike, low ped. ................................ ............... 74 A 2 Base, pre timed, medium veh, high bike, high ped. ................................ ............ 75 A 3 Base, pre timed, high veh, low bike, low ped. ................................ ..................... 76 A 4 Base, pre timed, high veh, high bike, high ped. ................................ .................. 77 A 5 Base, semi actuated, medium veh, low bike, low ped. ................................ ....... 78 A 6 Base, semi actuated, medium veh, high bike, high ped. ................................ .... 79 A 7 Base, semi actuated, high veh, low bike, low ped. ................................ ............. 80 A 8 Base, semi actuated, high veh, high bike, high ped. ................................ .......... 81 A 9 Pedestrian, pre timed, medium veh, low bike, low ped. ................................ ...... 83 A 10 Pedestrian, pre timed, medium veh, high bike, high ped. ................................ ... 84 A 11 Pedestrian, pre timed, high veh, low bike, low ped. ................................ ............ 85 A 12 Pedestrian, pre timed, high veh, high bike, high ped. ................................ ......... 86 A 13 Pedestrian, semi actuated, medium veh, low bike, low ped. .............................. 87 A 14 Pedestrian, semi actuated, me dium veh, high bike, high ped. ........................... 88 A 15 Pedestrian, semi actuated, high veh, low bike, low ped. ................................ .... 89 A 16 Pedestrian, semi actuated, high veh, high bike, high ped. ................................ 90 A 17 Bicyclist, pre timed, medium veh, low bike, low ped. ................................ .......... 92 A 18 Bicyclist, pre timed, medium veh, high bike, high ped. ................................ ....... 93

PAGE 8

8 A 19 Bicyclist, pre timed, high veh, low bike, low ped. ................................ ................ 94 A 20 Bicyclist, pre timed, high veh, high bike, high ped. ................................ ............. 95 A 21 Bicyclist, semi actuated, medium veh, low bike, low ped. ................................ .. 96 A 22 Bicyclist, semi actuated, med ium veh, high bike, high ped. ................................ 97 A 23 Bicyclist, semi actuated, high veh, low bike, low ped. ................................ ........ 98 A 24 Bicyclist, semi actuated, high veh, high bike, high ped. ................................ ...... 9 9 A 25 Combination, pre timed, medium veh, low bike, low ped. ................................ 101 A 26 Combination, pre timed, medium veh, high bike, high ped. .............................. 102 A 27 Combination, pre timed, high veh, low bike, low ped. ................................ ....... 103 A 28 Combination, pre timed, high veh, high bike, high ped. ................................ .... 104 A 29 Combination, semi actuated, medium veh, low bike, low ped. ......................... 105 A 30 Combination, semi actuated, medium veh, high bike, high ped. ...................... 106 A 31 Combination, semi actuated, high veh, low bike, low ped. ............................... 107 A 32 Combination, semi actuated, high veh, high bike, high ped. ............................ 108 A 33 Alternative, pre timed, medium veh, low bike, low ped. ................................ .... 110 A 34 Alternative, pre timed, medium veh, high bike, high ped. ................................ 111 A 35 Alternative, pre timed, high veh, low bike, low ped. ................................ .......... 112 A 36 Alternative, pre timed, high veh, high bike, high ped. ................................ ....... 113 A 37 Alternative, semi actuated, medium veh, low bike, low ped. ............................ 114 A 38 Alternative, semi actuated, medium veh, high bike, high ped. .......................... 115 A 39 Alternative, semi actuated, high veh, low bike, low ped. ................................ .. 116 A 40 Alternative, semi actuated, high veh, high bike, high ped. ................................ 117

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Sun glare on a traffic camera in Pinellas County, Florida. ................................ .. 20 2 2 Types of bicycle paths. ................................ ................................ ....................... 21 2 3 Phasing scheme to accommodate bicyclists and pedestrians. ........................... 22 2 4 Average user delay (s) for current versus alternative plan in Boston crossing. .. 23 2 5 Overhead view of influence area of pedestrian detection technology. ................ 25 2 6 Types of overhead detection for vehicles. ................................ .......................... 27 2 7 Strengths and weaknesses of detection technologies for vehicles. .................... 28 2 8 Series of different detect ion technologies for bicyclists ................................ ..... 29 2 9 Iteris PedTrax video detection. ................................ ................................ ........... 30 2 10 The CITIX IR overhead pedestrian detection device. ................................ ......... 30 2 11 Dutch style intersection designed to accommodate bicyclists safely. ................. 31 2 12 Two stage turn queue box. ................................ ................................ ................. 32 2 13 Effect of curb radii and parking on right turning paths ................................ ....... 32 2 14 Walking distance versus curb radius. ................................ ................................ 33 3 1 Base intersection. ................................ ................................ ............................... 36 3 2 Pedestrian intersection. ................................ ................................ ...................... 37 3 3 Bicyclist intersection. ................................ ................................ .......................... 38 3 4 Combined feature intersection. ................................ ................................ ........... 39 3 5 Staggered pelican crossing. ................................ ................................ ............... 40 3 6 Alternate intersection. ................................ ................................ ......................... 40 3 7 Base lane configuration. ................................ ................................ ..................... 43 4 1 Base design vs. pedestrian design individual weighted delays. ....................... 49 4 2 Base design vs. bicyclist design individual weighted delays. ........................... 50

PAGE 10

10 4 3 Base design vs. combination design individual weighted delays ..................... 51 4 4 Base design vs. alternative design individual weighted delays. ....................... 51 4 5 Combined weighted delay base vs. pedestrian. ................................ .............. 52 4 6 Combined weighted delay base vs. bicyclist. ................................ ................... 53 4 7 Combined weighted delay base vs. combination. ................................ ............ 53 4 8 Combined weighted delay base vs. alterna tive. ................................ ............... 54 4 9 Total vehicle delay base vs. pedestrian. ................................ .......................... 55 4 10 Total vehicle delay base vs. bicyclist. ................................ .............................. 56 4 11 Total vehicle delay base vs. combination. ................................ ....................... 56 4 12 Total vehicle delay base vs. alternative. ................................ .......................... 57 4 13 Base design vs. pedestrian design queue changes pre timed. ..................... 58 4 14 Base design vs. pedestrian design queue changes semi actuated. ............. 59 4 15 Base design vs. bicyclist design queue changes pre timed. ......................... 60 4 16 Base design vs. bicyclist design queue changes semi actuated. ................. 60 4 17 Base design vs. combination design queue changes pre timed. .................. 61 4 18 Base design vs. combination design queue changes semi actuated. ........... 61 4 19 Base design vs. alternative design queue changes pre timed. ..................... 62 4 20 Base design vs. alternative design queue changes semi actuated. ............. 63 4 21 Vehicle average speed fluctuations base vs. pedestrian. ................................ 64 4 22 Vehicle average speed fluctuations base vs. bicyclist. ................................ .... 65 4 23 Vehicle average speed fluctuations base vs. combination. ............................. 65 4 24 Vehicle average speed f luctuations base vs. alternative. ................................ 66 4 25 Alternative intersection right turn queueing. ................................ ....................... 67 5 1 Base configuration right turn issue. ................................ ................................ .... 69 A 1 Base configuration on VISSIM as pre timed. ................................ ...................... 82

PAGE 11

11 A 2 Base configuration on VISSIM as semi actuated. ................................ ............... 82 A 3 Pedestrian configuration on VISSIM as pre timed. ................................ ............. 91 A 4 Pedestrian configuration on VISSIM as semi actuated. ................................ ...... 91 A 5 Bicyclist configuration on vissim as pre timed. ................................ ................. 100 A 6 Bicyclist configuration on vissim as sem i actuated. ................................ .......... 100 A 7 Combination configuration on VISSIM as pre timed. ................................ ........ 109 A 8 Combination configuration on VISSIM as semi actuated. ................................ 109 A 9 Alternative configuration on VISSIM for pre timed. ................................ ........... 118 A 10 Alternative configuration on VISSIM for semi actuated. ................................ ... 118

PAGE 12

12 LIST OF ABBREVIATIONS AASHTO American Association of State and Highway Transportation Officials ALT Alternative intersection BASE Base intersection BIKE Bicyclist intersection COMBO Combination intersection HB High volume bicyclists HCM Highway Capacity Manual HP High volume pedestrians HV High volume vehicles LB Low volume bicyclists LP Low volume pedestrians MV Medium volume vehicles PED Pedestrian intersection PRET Pre timed signal system SEMI Semi actuated signal system

PAGE 13

13 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 Engineering OPERATIONAL EFFECTS TO DIFFERENT TRANSPORTATION MODES AT SIGNALIZED INTERSECTIONS FROM DIFFERING GEOMETRIES, SIGNAL SYSTEMS, AN D VOLUME LEVELS By Tyler J Valila December 2017 Chair: Lily Elefteriadou Major: Civil Engineering In the United States, walking and bicycling have become larger shares of transportation. Signalized intersections have not changed to a high degree geometrically or technologically to accommodate larger volumes of these modes in the presence of motor vehic les. The most advanced signal systems used today do not take pedestrians or bicyclists into account as being equal to the motor vehicle. Simple steps can be taken at the signalized intersection to improve serviceability for all users. The first objective o f this research is to identify safe designs for pedestrians and bicyclists at signalized intersection. The second objective is to create and simulate these intersection characteristics with altering geometries, signal systems and v olumes to determine which combinations produce re duction in delay The third objective is to create a set of guidelines based on the results that summarizes these operational effects. This research is meant to take an in depth look at design considerations and resulting operational effects that an engineer will need to consider for in tersection betterment projects The methodology involves building common scenarios seen in a signalized intersection using a micro simulator and finding which designs can improve

PAGE 14

14 intersection functionality the best. All results were yielded from the micro simulator VISSIM. Measures of effectiveness include delay, vehicle queueing, and average vehicle speed. These measurements were compared to base data to look for relative impacts. The results show good improvements are possible to an intersection with simple modifications. Implementing set back cross walks can reduce vehicle delay. Downstream crossing points for bicyclists can increas e vehicle average speed. A combination of these features gives favorable results for weighted mode delay. Finally, staggered crosswalks at intersections give little to no operational benefits. This research was performed solely in simulation and should be confirmed with real world field results.

PAGE 15

15 CHAPTER 1 INTRODUCTION Background The United States (US) has a car centric society that permeates into signal operations across the country. Only recently have attempts been made to give bicyclists shared or indivi dual lanes. Pedestrian crossings are utilized in intersections at the expense of vehicle delay. In a new future of transportation growth and innovation, these three mode choices will need molding to create next generation intersections that can reduce dela y and increase safety. Through advanced signal phasing plans, new signal systems, and effective geometric layouts, the new American intersection can strengthen the transportation network and elevate US guidelines. Many s tandards exist for differ ent geometric layouts that affect operations. Many signalized intersections now incorporate bike lanes in the pavement or sidewalk as a shared or singular lane. The US is experiencing rising bicycle usage for commutes to work with a 60.8% increase from 2000 to 2010 ( McKenzie 2014 ) Many communities in the US have high pedestrian and bicyclist traffic and are in need of innovative ideas. Installations of bicyclist friendly intersections are a reflection of this growth. In these intersections, bicyclists and pedestrians have more advantages. Standard signal systems hav e fixed pedestrian signal times with phases pre determined Adaptive signal systems have reduced delay for vehicles in many scenarios but do not account for pedestrians or bicyclists Attempts at creating adaptive signals that accommodate these modes are in the development stages. Preferred phasing plans for bicyclist and pedestrian movements typically implement concurrent phases parallel to vehicle movements.

PAGE 16

16 Detection systems have advanced enough to detect individual pedestrians and bicyclists with direction of travel. These detection systems for pedestrians and bicyclists are no t typically used in conjunction with vehicle signals to enhance operations Research Objectives The objectives of this research are to : 1. Identify s afe designs for pedestrians and bicyclists at signalized intersections 2. Create and simulate signalized intersection s selected from objective 1 with alternating combinations of geometric characteristics and phasing patterns to determine which combinations produce optimal changes in delay for all modes, queueing for vehicles, and speed for vehicles. 3. Create a set of guidelines based on the summarized conclusions for the operational effects of different scenarios Document Organization The literature review in C hapter 2 will present an in depth look at current technologies and ideas such as signal systems and designs that can help advance this research further. Chapter 3 reviews methodology and the simulator capabilities Chapter 4 contains results and explain s the trends of the data. Chapter 5 c ontains conclusions and recommendations drawn from the results to determine what designs work best to enhance operational effects at the signalized intersection.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW Chapter 2 describes current signal technology and detection for vehicles, bicyclists, and pedestrians with critiques of intersections accommodating all modes. Chapter 2 also discusses effective geometric layouts and phasing patterns for an inte rsection to accommodate all modes. Examples of current systems around the world that effectiv ely combine modes are discussed. Adaptive signals are not used in these analyses but in the future can be used to further similar research. Adaptive signals are discussed to outline where the technology could improve to better intersections in the future in conjunction with findings herein. Adaptive signals have existed for decades and have been exclusive to motor vehicles. Technologie s that can detect bicyclists and pedestrians either are beginning to be implemented or are being designed. The intersection itself has undergone changes in recent years in the United States including more bicyclist friendly layouts Pedestrian crossings ha ve not changed by any appreciable degree. To ensure safe and effective future intersections, the outlined modes will need sensible systems and ideal geometry. Signal Control and Phasing The primary types of signal control in the US is pre timed, semi actu ated, fully actuated and adaptive ; each have benefits and dis benefits. Vehicles Pre timed signals have green times that are fixed, based on historical data Semi actuated signals allow for a major road to continuously flow with permitted lefts until a side street sends a call for a phase. Pre timed and Semi actuated signals are popular at isolated intersections Fully actuated signals utilize minimum and maximum

PAGE 18

18 greens for all approaches with detection on all approach es. Many types of adaptive signal systems have been developed. Adaptive signals are installed at 3% of the nation's traffic signals ( Sanburn 2015 ). The elements of an adaptive signal control system include utilizing different detection methods to feed algorithms that have the goal of increasing capacity of an intersection and reducing delay in real time based on existing traffic ( Gettman et al. 2013 ). The five primary adaptive signal systems are SCOOT, SCATS, OPAC, RHODES, and ACS Lite. SCOOT stands for the split cycle offset optimization technique and is a popular system in the world developed by the Transport Research Laboratory in the United Kingdom. It works by dividing a network into regions with nodes meant to represent signalized i ntersections. The three optimizers in SCOOT are the split, offset, and cycle time. Based on prevailing conditions, the system will slowly adjust the signal timing plan. Zhao and Tian ( 2012 ) wrote that t he system could maintain a const ant coordination of the entire signal network SCOOT systems take traffic data in real time and works to minimize wasted green time at any intersection while reducing delays by synchronizing adjacent signal phases (Elefteriadou et al. 2015 ). SCATS stands for the Sydney coordinated adaptive traffic system and is a popular system in the world, which has been installed in 27 countries ( Elefteriadou et al. 2015 ). The system was developed by the Roads and Traffic Authority of New South Wales i n Australia. SCATS has three control levels represented by central, regional, and local levels. Each intersection distributes computations between a regional computer and a field controller, and over time can optimize itself Zhao and Tian ( 2012 ) say that t his system can be utilized in time of week coordination, individual

PAGE 19

19 intersections, and more SCATS limitations include not providing arrival prediction, queue estimation, or phase sequence optimization. So, it only acts as a regular signal system bu t is fully actuated. OPAC stands for optimized policies for adaptive control and features an algorithm that calculates signal timings to minimize total intersection delays and stops OPAC was the first demand responsive signal control. Developed at the Uni versity of Massachusetts Lowell, this system can make decisions on whether or not a phase should run ( Zhao and Tian 2012 ). According to the developers, it provides better results than off line systems. Unfortunately, due to technology at the time, the approach could not be implemented in real time. Numerous versions of OPAC were released in subsequent years and can optimize up to eight phases within the ring and barrier configuration along with lead/lag left turns ( Elefteriadou et al. 2015 ). RHODES stands for real time hierarchical optimized distributed effective system and was developed at the University of Arizona in the 1990's RHODES can take in data from multiple sensor types and generate optimal signal control plans. The system produces real time traffic flow predictions o n a corridor and optimize s th e flow with phase timing. Utilization of prediction techniques for individual cars and platoons are considered for the system ( Zhao and Tian 2012 ). RHODES uses a hierarchy structure that ranks from highest to lowest dynamic network loading which captures characteristics of geometry and route selection of travelers. Then network flow control allocates green times made by the decisions. There is inte rsection control that selects individual phases from observed and predicted volumes ( Elefteriadou et al. 2015 ).

PAGE 20

20 ACS Lite stands for adaptive control software lite and is sponsored by the Federal Highway Administration (FWHA), which similar features to th e other systems but also boasts low cost, compatibility with many current sensor networks, and simple calibration ( Zhao and Tian 2012 ). ACS Lite can control up to 16 consecutive intersections in a loop ( Elefteriadou et al. 2 015 ). This system's primar y goal is to be adaptive to traffic changes but maintain time of day schedules This system is closest in similarity to a regular non adaptive signal and has low incremental cost of installation. The system adjusts splits every 5 to 10 minutes, a much high er rate than other systems described. This system was designed to minimize operation and maintenance costs, and improvements at intersections are modest ( Elefteriadou et al. 2 015 ). Rhythm Engineering has been implementing a new adaptive control system called InSync InSync works by using cameras to identify vehicles in queue and assign A drawback of the cameras is weather related events including the suns glare as shown in Figure 2 1 from the Pinellas County Traffic Management Center (PCTMC). The suns glare can make the system believe vehicles are present that are not and vice versa. Figure 2 1. Sun glare on a traffic camera in Pinellas County, Florida ( image c ourtesy of PCTMC).

PAGE 21

21 In the bottom right quadrant of Figure 2 1 the suns glare makes the camera believe a vehicle is in queue when no vehicle is present. This issue is not exclusive to adaptive signal systems. There are no set cycle lengths, transition periods, or phasing pat terns ( Elefteriadou et al. 2 015 ). The system converts analog control to digital control, optimizes the individual intersection, and then optimizes the total system of intersections. O ther systems do not go to this level of detail. Some issues with camera d etection and software persist but the company has worked out most problems. Rhythm has installed InSync at more than 2000 intersections across 31 states ( Rhythm 2017 ) Th e University of Florida Transportation Institute has studied the before and after implementations of adaptive deployments across Florida. Results have shown modest decreases in travel time per mile across all InSync deployments studied with the exception of heavily congested corridors ( Elefteriadou et al. 2 015 ). Bicyclists Signal technology for bicyclists is behind the technology used for vehicles. Adaptive signals for bicyclists have not been found in literature but InSync is expecting to release a feature that includes minimum green timing for bicyclists in corresponding through movements. Bicycle lanes are categorized in the HCM as shown in Figure 2 2 Figure 2 2 Types of bicycle paths ( reprinted from TRB 2010 )

PAGE 22

22 In the US bicycle lanes are becoming more commonplace. Most of the types of bike paths shown in F igure 2 2 coincide with vehicles in the immediate vicinity. B icycle usage is rising in the U nited States and therefore the issue of accommodating bicycles in intersections will continue. Copenhagen, Denmark sees volumes exceeding 30,000 bicycles per day in both directions on spring and fall days with good weather ( TRB 2010 ). This number is much higher than virtually any American city corridor and explains why more measures are taken in European centers When bicyclists are present at an intersection, the users typically get a sole phase for all movements or concurrent and pa rallel phases with vehicle movements. Sole phases reduce an intersections capacity and increase cycle length. All pedestrian phases are usually necessary within college campuses and similar areas with remarkably high pedestrian volumes. An advantage to pro tected yet concurrent phasing is most helpful when the stop line for a bike is further downstream than the vehicle stop bar. One phasing scheme that accommodates bicyclists and pedestrians with conflicting right turns is in Figure 2 3 Figure 2 3 Phasing scheme to accommodate bicyclists and pedestrians ( reprinted from Furth et al. 2014 ).

PAGE 23

23 Figure 2 3 shows that bicyclists move with the flow of traffic for this particular phasing plan. A study on European intersections ha s found some intersections utilize leading bicycle intervals ( Gilpin et al. 2015 ) A study of Roxbury Crossing in Boston was completed comparing the current all pedestrian phase with a concurrent but protected phase with results in Figure 2 4 The intersection has high pedestrian volumes due to its proximity to schools. At this particular intersection, bicyclists can be unwilling to wait for an all pedestrian phase and proceed without any protection. The following protected but concurrent phasing can help optimize efficiency of time for pedestrians, bicyclists, and vehicles. Giving pedestrians and bicyclists protected increases safety for all. Figure 2 4 Average user delay (s) for current versus alternative plan in Boston crossing ( reprinted from Furth et al. 2014 ) The data shows that changing from an all pedestrian and bicyclist phase to concurrent phases can significan tly reduce delay for bicyclis ts and pedestrians. Within the US no national guidelines exist for permitted conflicts at bike crossings ( Furth et al. 2014 ). Dutch stop bars for bicyclists are between 40 and 60 feet downstream and have their own phases ( Furth et al. 2014 ). These bicyclist phases may begin earlier or later than the vehicular phase to allow more time to reduce conflicting right turns ( Furth et al. 2014 ).

PAGE 24

24 Pedestrians Pedestrians currently cross intersection s with pre set crossing times T here are no implemented adaptive signals for pedestrian usage. Some areas such as Ithaca, NY and Athens, OH have more than 35% of commuters walking to work but the average for the country is around 3% ( McKenzie 2014 ). With some areas having such large pedestrian volumes, a change in technology could help enhance opera tions. Some intersections allow through movements of vehicles to run at the same time as parallel pedestrian crossing s, similar to bicyclists This works if the signals are pre timed to allow enough time for a pedestrian to clea r the intersection. The length of time it takes a fixed group of pedestrians to cross an intersection does not change, so running pedestrians parallel to actuated or adaptive signals for vehicles would not always work effectively If a through movement req uires less green time then pedestrians need to clear an intersection, the efficien cies lessened There are cases where pedestrians are given pre timed phase s in an adaptive signal intersection and the benefits of the adaptive signal reduced yet scheme can be more efficient than giving pedestrians and bicyclists an individual phase ( Furth et al. 2014 ) whic h is similar to bicyclist findings A thesis by Hu (2014) describes the problem with pedestrians in adaptive intersections The thesis involved simulated movements using real data of pedestrian and traffic volumes in synchro. (2014) findings were the following: 1. Pedestrian actuations can increase the control delay both under TOD plan and ASCT plan. 2. The impact of pedestrian activities is more significant on the ASCT system than the TOD coordination. 3. Pedestrian activities offset some of the benefits on delays brought by ASCT.

PAGE 25

25 4. ASCT reduces delay when compared to a TOD plan even when pedestrian activities increase delay. The study also gave recommendations for future ASCT projects on urban roads. The delays caused by pedestrian activities should be considered during both the design and planning phases. A recommendation for future research includes incorporating new detection devices to determine durations and sequence of pedestrian intervals without compromising safety. A research team in C hina developed an adaptive pedestrian crossing signal control system. The developed system was able to effectively identify the pedestrian waiting area and calculate pedestrian wait time and total pedestrians ( Xiao et al. 2013 ). According to the research team, the video sampling device is fitted on a stand on the side of the crossing and the video sequences are taken in real time ( Xiao et al. 2013 ) The control unit calculates the amount of pedestrians and their respective waiting time. If the amount of pedestrians or waiting time goes beyond a certa in limit, a crossing signal is activated by the control unit and displayed by the display unit. The video unit in the study scanned the area surrounded by a black rectangle in Figure 2 5 Figure 2 5 Overhead view of influence area of pedestrian detection technology ( reprinted from Xiao et al. 2013 )

PAGE 26

26 The proposed background algorithm follows seven ( 7 ) steps that take the intensities of pedestrian volumes and determine when a green signal is warranted To test the validity of the algorithms, most of the simulations were completed at T junctions without traffic signals. The results showed hardware co mponents of adaptive pedestrian signals are viable, the algorithm used can build the background reconstruction well, and image sequence processing is effective. The researchers note d that future work needs to be done with experime nts that are more realistic Currently at signalized intersections, pedestrians either have one entire phase or run parallel to vehicle movements The problem with one phase is it reduces the effectiveness of the adaptive signal for vehicles. The problem with pedestrians moving parallel to adaptive signals when green is when the minimum time required to clear is greater than time allotted for vehic ular traffi c. This causes unnecessary delay for vehicles on the opposing approaches. Coordination can be lost if the pedestrian timing exceeds allocated time for the corresponding vehicle movement. InSync has deployed time of day pedestrian exclusive phase s for school release. Detection Detection technology is rapidly advancing for all modes of transportation. The purpose for including detection tech nologies is to show advancements in the field especially for pedestrians and bicyclists. This information c an be useful for future research when determining efficient ways to calculate crossing times All actuated and adaptive signal systems rely on detection to make algorithms work. Vehicles have overhead cameras, radar, loop detectors, pressure plates, and m ore. Bicyclists use loop detectors, overhead and side detection, and push buttons. Pedestrians typically only have push buttons but many companies are developing both

PAGE 27

27 overhead and side detection systems that can accurately count the number of pedestrians e ntering a crosswalk area. Vehicles Vehicle sensors are either in roadway or over roadway. In roadway sensors include loop detectors, weigh in motion sensors, magnetometers, tape switches, microloops, pneumatic road tubes, piezoelectric cables ( Mimbela and Klein 2007 ). These are installed either in the pavement, subgrade, and attached to the road surface. These do require disruption to traffic flow to install and maintain. Resurfacing work can create the need to re install these technologies as well ( Mimbela and Klein 2007 ). Over road way sensors include video image processors, microwave radar sensors, ultrasonic, passive infrared, and laser radar sensors. Passive acoustic sensors can be used for adjacent roadway sensing ( Mimbela and Klein 200 7 ). Some over roadway technologies are outlined in Figure 2 6 Figure 2 6 Types of overhead detection for vehicles ( reprinted from Mimbela and Klein 2007 )

PAGE 28

28 These overhead detection devices are commonplace in intersections in the US. A comparison between strengths and weaknesses for different sensor technologies is presented in Figure 2 7 Figure 2 7 Strengths and weaknesses of detection technologies for vehicles ( reprinted from Mimbela and Klein 2007 ) Some agencies prefer radar to cameras due to natural events such as the suns glare Bicyclists Many companies have developed products capable of accurately counting bicyclists in urban environments. These real time detection systems tech nologies range from tubes to in ground sensors to post mounted sensors ( Eco Counter 2017 ). Eco Counter is a company out of Canada that specializes in these types of detection systems. One detection system called MULTI can count both pedestrians and bicyclists

PAGE 29

29 at the same time accurately at distances of up to 6 meters. Another technology called the ZELT Loop ( Figure 2 8 ) can detect direction and speed of traveling bicyclists on shared roads. A B Figure 2 8 Series of d ifferent detection technologies for bicyclists A) Sketch of sublayer ZELT installatio n, B) Bicycle detection system on shared road ( reprinted from Eco Counter 2017 ) These technologies i ntegrate well into communities and have an aesthetic touch while serving an engineering purpose. They also can accurately tell which way a bike is heading. Pedestrians Pedestrian detection systems generally rely on mounted detection devices. Many companies have developed detection systems use d for a wide variety of implementations. Iteris is a company from California that develop s pedestrian detection systems. directional counting and speed tracking of pedestrians within the crosswalk ( Iteris 2017 ) A

PAGE 30

30 camera view of a pedestrian crossing is shown in Figure 2 9 The company Eco Counter sells the CITIX IR camera ( Figure 2 10 ) that counts pedestrian movements overhead over a sidewalk. Figure 2 9 Iteris PedTrax video detection ( reprinted from Iteris 2017 ) The CITIX IR Eco Counter 2017 ) A B Figure 2 10 The CITIX IR overhead pedestrian detection device shown as A) From the pedestrian view, B) Sketch of the technology ( reprinted from Eco Counter 2017 )

PAGE 31

31 The system can indicate direction of travel as well. Standard equations exist that take crosswalk length, width, and pedestrian volumes into account for a preset clearance time. If the amount of pedestrians are detected, the time can adjust to reduce dela y. However, this is not current practice Intersection Design In the United States, bicyclist friendly intersections are installed in some locations. The bicyclist friendly intersection is a viable option to accommodate vehicles, bicyclists, and pedestrians safely ( Gilpin et al. 2015 ). Figure 2 11 in Davis, CA shows a Dutch style protected intersection designed for bicyclist safety. Figure 2 11 Dutch style intersection designed to accommodate bicyclists safely ( reprinted from Anders e n 2015 ) Many intersections utilize this type of design involving downstream and offset bicyclist crossings that can help bicyclists stay protected. Bumping out corners can reduce the turning time of vehicles to reach bicyclists in crosswalks and increases safety. The two stage turn queue box lets bicyclists complete safe left turns with less conflicts at the e xpense of higher delay ( NACTO 2011 ) as shown in Figure 2 12

PAGE 32

32 Figure 2 12 Two stage turn queue box ( reprinted from NACTO 2011 ) A report in 2003 for the FHWA outlined signalized intersection safety in Europe. One method used by the Dutch to improve intersection safety is to install grade separated crossings for bicyclists and pedestrians ( Fong et al. 2003 ). AASHTO guide lines on intersection geometry states that for vehicle operations should be balanced against the needs of pedestrians and the difficulty of acquiring additional right of ( AASHTO 2012 ). Crosswalk distances change along with adjustments to curb radiu s as shown in Figure 2 13 A B Figure 2 13 Effect of curb radii and parking on right turning paths for A) 15 foot curb radius, B) 25 foot curb radius ( reprinted from AASHTO 2012 )

PAGE 33

33 The difference in curb radii makes a distinct change in movement capability for vehicles. Figure 2 14 shows how changes in curb radii affect length of crosswalks. Figure 2 14 Walking distance versus curb radius ( reprinted from AASHTO 2012 ) AASHTO (2012) recommends that the curb radii should be coordinated with crosswalk distances or special designs should be used to make crosswalks efficient for all pedestrians For large radiuses of 40 feet or more (typical fo r refuge islands), tapers should be provided to fit paths of large trucks or buses. Setback crosswalks allow for pedestrians to be removed from vehicular traffic movements conflicts Curb parking lanes and restrictive parking can increase the usable radius. Proper channelization can increase capacity for pedestrian and enhanced guidance for motorists ( AASHTO 2012 ). Mitigating the right turn bicycle and pedestrian conflicts through pocket bik e lanes, and raised crossings like in Boulder, CO can incre ase safety ( Furth et al. 2014 ).

PAGE 34

34 Summary Adaptive signal technology has been effective in reducing delay for motorists but does not accommodate bicyclists or pedestrians well. Pedestrian and bicycle phases can reduce benefits from adaptive signal installation. Detection methods are present and sophisticated for both pedestrians and bicyclists, but no adaptive signal system has implemented these detections into algorithms for all modes. Intersection geometry has fundamentally c hang ed with the incorporation of bicycle lanes Changes to curb radii can affect the travel distance for pedestrians and bicyclists across roads but comes at the cost of motorist comfort and right of way. The two most prevalent safety enhancements that wil l be used for analysis is the setback crosswalk and the downstream bicyclist crossing. In order to enhance operational effects for vehicles, pedestrians, and bicyclists at signalized intersections, a combination of phasing patterns, intersection geometry, and detection technology can be implemented.

PAGE 35

35 CHAPTER 3 M ETHODOLOGY Chapter 3 will review intersection configurations used for testing, the simulator software, scenarios to be included, and how data results are analyzed. The objectives of this research are to identify safe designs, simulate the designs, and then determine which work best for certain situations. To accomplish this, safe designs from literature review are simulated in VIS SIM. Intersection Configuration Before beginning the analysi s, safe designs and features of intersections must be incorporated into the different scenarios. The configurations are base design pedestrian design, bicycl ist design, combin ation ( bike/ped ) design and an alternative design All scenarios will have the same lane geometry. All intersections will share the following characteristics: No coordination with any other inter section Zero grade on all approaches 1 2 foot Lanes 2% Heavy vehicles for all turning movements 10 foot Wide crosswalks Stop bars in line with each other by approach Good sight distance (no buildings or objects in the drivers way) A 40 MPH speed limit for the major road 25 MPH speed limit for the minor road Left turn and right turn bays of 100 feet For all scenario s pedestrians and bikes have concurrent phas ing with vehicle movements. B ike movements for the base configuration are in line with the vehicle lanes as shown in Figure 3 1 Bike movements follow the original direction of travel L eft and right turns will not be simulated The downstream and perpendicular distance from the vehicle stop bar to the parallel pedestrian crosswalk is 20 feet and 1 0 feet

PAGE 36

36 respectively with an exception for s outhbound through/right movements due to lane geometry Figure 3 1 Base intersection The base intersection is the source of comparison for all other scenarios. For t he pedestrian intersection, A downstream crossing point of 40 feet and perpendicular distance of 3 0 feet to a crosswalk is used for the design. Th e turn radius is small enough to reduce vehicle speed. This design utilizes setback c rosswalks shown in Figure 3 2 Vehicles and bicyclists have to cross a further distance to clear. The 40 foot measurement is a typical lower side measurement for Dutch intersections that use this kind of geom etry and will allow space for right turning vehicles to queue without impeding through movements Pedestrians are given a head start for

PAGE 37

37 crossing the street This scenario will not reduce the length that a pedestrian must cross, as no lanes are eliminated. Figure 3 2 Pedestrian intersection Pedestrians have a downstream starting point relative to vehicles. This coupled with the head start timing allow s pedestrians to get out of a vehicles way quicker, and allow s right turning vehicles to not obstruct through movements. This design can effect sight distance for vehicles The bicyclist intersection has through and right moving bikes in exclusive lane s next to the pedestrian sidewalk A bike ramp from the road to the sidewalk is used This feature is shown in Figure 3 3

PAGE 38

38 Figure 3 3 Bicyclist intersection The bicyclist design re duces the proximity of a bicyclist to a motor vehicle by shifting the crossing point. In a paper by Stanek and Alexander ( 2015 ) tests were performed using VISSIM to determine the operational effects to vehicles when placing a bicyclist crossing downstream of the vehicle stop bar. There are four scenarios identified in the research for when bicyclists and vehicles interact which are right turning drivers yielding, bicyclists getting a lead interval, bicyclists getting separated and protected phasing, and right turning cars move unimpeded while bikes move with through traffic ( Stanek and Alexander 2015 ) The bicyclist design in Fi gure 3 3 has downstream crossings for bicyc lists that for the westbound approach utilize s the

PAGE 39

39 authors fourth method. Bicyclists have the same signal timi ngs as vehicles in this design. The combined intersection utilize s the features of both the pedestrian and bicyclist friendly intersection together The combined intersection is shown in Figure 3 4 Figure 3 4 Combined feature intersection The setback crosswalks coupled with shifted bicyclist crossing points should allow through moving vehicles in shared right turn lanes to move unimpeded. An alternate intersection is also teste d This intersection design is based on the staggered pelican crossing ( Department for Transport 2017 ) used in the United Kingdom An example of a staggered pelican crossing is shown in Figure 3 5

PAGE 40

40 Figure 3 5 Staggered pelican crossing ( reprinted from Department for Transport 2017 ) The alternate design (F igure 3 6) utilizes the concept of the staggered pelican crossing but at a signalized intersection crossing This allows right turning vehicles ample queue storage to allow through movements to proceed uninhibited. Figure 3 6 Alternate intersection

PAGE 41

41 The staggered pelican requires two separate crossing times, but this alternate design utilizes one crossing time with an advanced pedestrian start. In a real world situation i f a pedestrian is unable to clear the intersection, they can use the island as refuge until the next crossing available. The alternate design also utilizes a median refuge isl and in conjunction with the staggered crosswalk Not pictured in F igure 3 6 is the required median space for the design The design would make more sense at an intersection where all approaches have medians already in place Bicyclists will not use this crosswalk feature. One benefit of this design is better sight distance for vehicles compared to the pedestrian or combination design. Simulation The program used to simulate the criteria is the micro simulator VISSIM. The program has many capabilities that are useful for this analysis. VISSIM capabilities include displaying different stop bars for diff erent modes to cross the street, the use of external controllers, all types of signal systems, determining queues of all modes and accurate representations of all modes and respective characteristics Downsides to the program VISSIM are a steep learning curve and the program being a blank canvas where minor details can be over looked Dynamic crossing times similar to discussion in literature review can be coded into VISSIM but would be out of sco pe for this research. Challenges from using VISSIM began with learning how to properly incorporate pedestrians and bicyclists. These types of modes require separate and models in the simulator. Vehicles use wiedemann 74 bicyclists use w iedemann 99 and pedestrians use a social force model. The volumes could also be set to stochastic or fixed. Fixed was chosen because a similar amount of volumes will occur between runs of a scenario. VISSIM requires the user to manua lly assign conflict areas where

PAGE 42

42 pedestrians have right of way over vehicles, opposing through moving vehicles have right of way over permitted left turns, and more. The user must also identify reduced speed areas for vehicles that clear the stop bar and are using re duced speed to complete a turn VISSIM defines delay of a mode as the optimal time subtracted from the actual time. Vehicle queues are measured simultaneously through the simulation regardless of whether the signal is green or red. For this research, queu es are defined in VISSIM as vehicles that are fully stopped The use of VISSIM lead to some issues in each configuration. The base configuration was the before moving onto other configura tions. Some issues were lining up opposing movements, using simple connections and making the correct modes appear in the correct lanes. One situation involved pedestrians traveling at the speed of vehicles using the vehicle lanes. No major issues arose with the pedestrian configuration or the bicyclist configuration. The combination configuration was very complex along with the alternative configuration. When moving the lanes to create the configurations, the conflict areas had to be re made and fitting the crosswalks into the medians required minor geo m etric adjustments Scenarios Different volumes of vehicles, bicyclists, and pedestrians are inter changed along with the signal system and configuration T he intersection s all have the same lane configuration as shown in Figure 3 7 regardless of signal system or geometry

PAGE 43

43 Figure 3 7 Base lane configuration D ifferentiating combinations of lane movements are used to get a better idea of how the geometry and signal system affect turning movements C ycle lengths and green time splits calculations are based on standard equations for the pre timed and semi actuated scenarios The optimal cycle length was used in nearly all pre timed scenarios with some exceptions where the minimum time was used. All pre timed scenarios have three phases beginning with eastbound and westbound left turns, then eastbound and westbound through and right turns, and finally all northbound and southbound movements. All semi actuated scenarios have two phases starting with all eastbound and westbound movements with permitted left turns and ending with all northbound and southbound movements. Using existing technology, c urrent equation s could determine required crossing time for pedestrians and bicyclists as they arri ve to cross the street This is different from current systems that have a pre set walk time fo r pedestrian s based on historical data Some intersections utilize

PAGE 44

44 through loop detector detection ( AASHTO Executive Commi ttee 2012 ). Th is technology is not used in these designs. For this research, a dynamic crosswalk timing is not used due to its complexity. Different volumes of pedestrians and bicyclists are simulated into the network with a n appropriately pre timed crossing time To determine the cross time necessary for the pedestrians E quation 3 1 ( TRB 2010 ) will be used. (3 1) Where Gp = minimum pedestrian green time in seconds 3.2 = pedestrian start up time in seconds L = crosswalk length in ft Sp = walking speed of pedestrians, usually taken as 3.5 ft/s Nped = number of pedestrians crossing during an interval WE = effective crosswalk width in ft Bicyclist crossing times are calculated from E quation 3 2 and 3 3 ( AASHTO Executive Committee 2012 ) (3 2) (3 3) Where, BCT STANDING = Bicycle crossing time (s) PRT = perception reaction time (1 sec)

PAGE 45

45 V = attained bike crossing speed (ft/s 14.7 ft/s for most people ) a = bike acceleration (1.5 ft/s 2 ) W = Intersection width (ft) L = typical bike length (6 ft) BMG = Bicycle minimum green time (s) Y = yellow change interval (s) R CLEAR = all red (s) The pre timed scenarios all have pedestrian crossing times (walk and do no t walk) that are equal to or less than the concurrent vehicle movements green time. The semi actuated signals have some situations where the minimum green time is increased for minor street movements to accommodate pedestrian arrivals. Minimum and pedestrian recalls are also implemented. The criteria used in VISSIM simulations is outlined in Table 3 1. Table 3 1 Criteria table for VISSIM analysis Geometric Design Signal System Vehicle Volume Ped/Bike Volume Base Pre Timed Medium Low Pedestrian Semi Actuated High High Bicyclist Combin ation Alternative These criteria are used for all scenarios and for comparison between each other. The vehicle volumes used reflect intermediate peak and peak period volumes while the pedestrian and bicyclist volumes will reflect an intersection with little demand versus high demand A summary of volumes for vehicles pedestrians and b icyclists is presented below in Table 3 2 and Table 3 3 :

PAGE 46

46 Table 3 2 Volume ta ble for vehicles (vehicle s per hour). Vehicle Movements Level NBL NBT NBR SBL SBT SBR EBL EBT EBR WBL WBT WBR Medium 20 120 50 40 300 85 100 880 30 60 350 30 High 24 144 60 44 330 94 140 1232 42 78 455 39 Table 3 3. Volume ta ble for pedestrians and bicyclists (individuals per hour). Pedestrian Movements Bicyclist Movements Level North South East West North South East West Low 100 100 50 50 50 50 40 40 High 400 400 200 200 200 200 160 160 Performance M easures Performance measures sought for analysis include: 1. The overall delay of left, through and right turning vehicles by approach 2. Bicyclist and p edestrian delay at cross ing points 3. The average vehicle speed of the system 4. Queue length changes on all vehicle lanes The program VISSIM produces reports on delay, queues, level of service (LOS), and average speeds. The HCM provides a separate methodology on calculating delay for pedestrians VISSIM resu lts are used for all analyse s Delay in the program is defined as the optimal travel time subtracted from the actual travel time. Queues in VISSIM are defined as any vehicle traveling at less than 0.1 mph effectively fully stopped VISSIM collects queue data continuously during an entire analysis period so queue results do not represent the average queue when a phase begins (HCM methodology) but rather at all times. Statistical Comparison Process The inte rsection s analysis are performed by comparing the base geometric layout results to different designs A co nfidence level of 95% is used for each scenario

PAGE 47

47 This yields N runs per scenario from E quation 3 4. There are 4 0 different scenarios (yielded from Table 3 1 ). (3 4) Where N = number of runs Std.S = Sample standard deviation from an initial ten runs (0.3 mph) t(1 a/2, N 1) = T score for 2 tail, 95% confidence, 9 degrees of freedom (2.2622) E = desired margin of error ( taken as 0.5 mph ) Equation 3 4 yields a required two runs per scenario but ten runs are used for all scenarios. T he sample collection sheet in Table 3 4 is used for inputting data for the appendix The collection sheet allows all required performance measures to be recorded in one sheet. One sheet is an average of ten runs and all sheets are included in appendix

PAGE 48

48 Table 3 4 Criteria table for comparative analysis Signal System Type x Vehicle Average Speed (mph) x Vehicle Volume Category x Weighted Veh Average Delay (s/veh) x Pedestrian Volume Category x Weighted Ped Average Delay (s/ped) x Bicyclist Volume Category x Weighted Bike Average Delay (s/bike) x Geometric Layout x Cycle Length (s) x Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL x x *NB East x NB x NBT x x NB West x SB x NBR x x SB East x EB x SBL x x SB West x WB x SBT x x EB North x SBR x x EB South x EBL x x WB North x EBT x x WB South x EBR x x WBL x x WBT x x WBR x x *NB East denotes NB travel on the eastern most crosswalk

PAGE 49

4 9 CHAPTER 4 RESULTS Chapter 4 reviews the results from simulations of all scenarios. This includes delay, vehicle queu eing, and vehicle average speed. First, the individual weighted delay of each mode are reviewed. Then, the weighted delay of all modes combined are reviewed. The final delay measurement will be total vehicle delay. The vehicle queue changes of all scenarios broken down by direction and approach are reviewed. Finally, vehicle average speed changes are compared. Delays The first measure of effectiveness is the delay of veh icles, bicyclists, and pedestrians at the intersection. Delay is a great indicator of gas usage with more delay totaling to more gas usage. Minimizing delay for all modes is imperative for an efficient intersection. Individual weighted delay is defined as the per person delay per mode. The total given in F igures 4 1 to 4 4 are sums of these individual delays and is not the weighted delay of all modes. The pedestrian designs delays are very consistent with the base design except for LP LB MV PRET whe re dela ys mostly decrease and LP LB HV SEMI where delays mostly increase ( F igure 4 1). Figure 4 1. Base design vs. pedestrian design individual weighted delays. -4 -3 -2 -1 0 1 2 3 4 5 6 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Change in Delay (sec) Scenario Individual Weighted Delay Pedestrian vs. Base Vehicle Pedestrian Bicyclist

PAGE 50

50 Delay goes down for vehicles in the pedestrian configuration because right turning vehicles are not immediately yielding to the crosswalk. The bicyclist configuration results in F igure 4 2 shows little to no improvement in any system. Bicyclists receive a downstream head start on vehicles with a small offset. This design h ad virtually no positive impact for bicyclists delay, but does give bicyclists a shorter crossing distance. Figure 4 2. Base design vs. bicyclist design individual weighted delays. When both the pedestrian and bicyclist features were combined, delay reduction was expected to be significant ( F igure 4 3) The results show t hat only one scenario, LP LB MV SEMI provides a reduction in all modes delay, albeit small. With such little interaction between vehicles, bicyclists, and pedestrians, this makes sense. The pre timed version did not have the same result most likely due to less time to cross per cycle. Bicyclist delay increased in the other seven scenarios. -2 -1 0 1 2 3 4 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Change in Delay (sec) Scenario Individual Weighted Delay Bicyclist vs. Base Vehicle Pedestrian Bicyclist

PAGE 51

51 Figu re 4 3. Base design vs. combination design individual weighted delays The alternate configuration has large increases in delay for pedestrians. This occurs because the crossing time required for pedestrians increases, which in turn reduce s the walk time allowed. The alternative layout also requires pedestrians to cross a longer distance than in other scenarios. The layout provides better sight distance to vehicles but as a whole does not help the system itself except for a slight bett erment in HP HB HV Semi actuated scenarios. Bicyclist delay remains close to the base data. Figure 4 4. Base design vs. alternative design individual weighted delays. -3 -2 -1 0 1 2 3 4 5 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Change in Delay (sec) Scenario Individual Weighted Delay Combination vs. Base Vehicle Pedestrian Bicyclist -2 -1 0 1 2 3 4 5 6 7 8 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Change in Delay (sec) Scenario Individual Weighted Delay Alternative vs. Base Vehicle Pedestrian Bicyclist

PAGE 52

52 The delay of the three modes are also combined into a wei ghted average of all users in the intersection. Figure 4 5 to 4 8 show the weighted delay for every mode combined The value was calculated by finding the sum of vehicle seconds delay, bicyclist seconds delay, and pedestrian seconds delay and dividing th e sum by the total users in the system. The pedestrian scenarios saw the best decrease in delay in the LP LB MV PRET scenario of 13 seconds to approximately 10.5 seconds This is because the scenario has the lowest possible traffic volumes combined with m ore phases in a pre timed setting. The pedestrian design helps w ith reducing weighted delay best overall. The combination design also has slight improvements as well due to the similar design. Figure 4 5 Combined weighted delay base vs. pedestrian. 8 10 12 14 16 18 20 22 24 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Weighted Delay of all modes (s) Scenario Combined Weighted Delay Pedestrian vs. Base Base Pedestrian

PAGE 53

53 The bicyclist weighted delay in F igure 4 6 shows expected results. Since the geometric changes are minimal, the delay of all modes is expected to remain close to the base. Figure 4 6 Combined weighted delay base vs. bicyclist. As expected, the combined weighted delay in F igure 4 7 follows a similar trend to the pedestrian delay. Figure 4 7 Combined weighted delay base vs. combination. 8 10 12 14 16 18 20 22 24 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Weighted Delay of all modes (s) Scenario Combined Weighted Delay Bicyclist vs. Base Base Bicyclist 8 10 12 14 16 18 20 22 24 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Weighted Delay of all modes (s) Scenario Combined Weighted Delay Combination vs. Base Base Combination

PAGE 54

54 The alternative configuration experiences rises in all scenarios with combined weighted delay. This configuration requires pedestrians to walk further distances than the base. The configuration is the same as the combination configuration with the e xception of a further pedestrian crossing distance, and a setback crosswalk. The rising effect is due to the large rises in pedestrian delay Figure 4 8 Combined weighted delay base vs. alternative. Weighted delay increases considerably for some scenarios in the alternative configuration. The alternative design reduces the available green time for pedestrians. If a design featured the alternative layout but bikes moved with pedestrians, the vehicle average speed could theoretically increase, vehicle delay would go down, but pedestrians and bicyc list would have to both cross further distances. The next comparison method for delay is the total vehicle delay of all motor vehicles. This measurement shows how many total seconds of waiting for motor vehicles are removed or gained from the system from geometric changes. 8 10 12 14 16 18 20 22 24 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Weighted Delay of all modes (s) Scenario Combined Weighted Delay Alternative vs. Base Base Alternative

PAGE 55

55 The m ost noticeable change is the p edestrian design in F igure 4 9. Every scenario sees a drop in delay of over 1,000 seconds each with a maximum of almost 6,000 seconds of delay removed from the system in the LP LB MV PRET scenario. Figure 4 9. Total vehicle delay base vs. pedestrian. The bicyclist design do es not show any noticeable improvements to the system for any scenario in F igure 4 10. Some scenarios see an increase in time such as the LP LB HV SEMI and the HP HP HV SEMI. These two scenarios share the same amount of vehicles and the same signal system, with the only change coming from the amount of pedestrians and bicyclists. This indicates that vehicles are sensitive to this geometric change in high vehicle volume environments. 10 12 14 16 18 20 22 24 26 28 30 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Total Delay (Thousands of sec) Scenario Total Vehicle Delay Pedestrian vs. Base Base Pedestrian

PAGE 56

56 Figure 4 10. Total vehicle delay base vs. bicyclist. Similar to other results in the delay measurements, the combin ation design performs well ( F igure 4 11) with every scenario seeing a drop in total vehicle delay The alternative design in F igure 4 12 contains only one scenario with a drop in total vehicle delay in LP LB MV SEMI. Figure 4 11. Total vehicle delay base vs. combination. 10 12 14 16 18 20 22 24 26 28 30 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Total Delay (Thousands of sec) Scenario Total Vehicle Delay Bicyclist vs. Base Base Bicyclist 10 12 14 16 18 20 22 24 26 28 30 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Total Delay (Thousands of sec) Scenario Total Vehicle Delay Combination vs. Base Base Combination

PAGE 57

57 Figure 4 12. Total vehicle delay base vs. alternative. The delay measure of effectiveness indicates that the pedestrian design works best overall, especially for pre timed signals. Delay decreases generally for all modes when set back crosswalks are implemented in configurations like pedestrian and combination. Designs with minimal geometric changes like the bicyclist configuration has virtually no change to the system. The alternati ve configuration in theory should see bicyclist and pedestrian waiting delay remain similar to the base but because less green time is allotted delay goes up. Vehicle Queues Queueing results are the second measure of effectiveness. When looking at the marginal results from changes in an intersection, the queue data can tell how many vehicles are waiting for their respective phase. Pedestrian and bicyclist queues are not used for comparison because vehicle queueing is considered to be more significant to 10 12 14 16 18 20 22 24 26 28 30 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Total Delay (Thousands of sec) Scenario Total Vehicle Delay Alternative vs. Base Base Alternative

PAGE 58

58 overall operational effects due to fuel usage and pedestrians and bikes moving as a group from queue where vehicles have to wait f or the downstream vehicle to move first The pedestrian configuration has little change in queueing in pre timed signals with the exception of the southbound through/right movement, eastbound left movement, and eastbound through/right movement in the HP H B MV scenario ( F igure 4 13) These movements contain the critical volumes for the three phases. In a maximum volume scenario, queue increases are expected The semi actuated signals for pedestrian design shows minim al change in queue ( F igure 4 14). Figure 4 13 Base design vs. pedestrian design queue changes pre timed. -20 0 20 40 60 80 100 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Pre timed Pedestrian vs. Base LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET

PAGE 59

59 Figure 4 14 Base design vs. pedestrian design queue changes semi actuated. The bicyclist configuration has many more movements seeing varying changes in queues than the pedestrian configuration. Sharp increases in northbound movements and southbound through/right movements persist for the LP LB HV PRET scenario as seen in F igure 4 15. This is likely because the minor street movements have a lower share of green time. The eastbound and westbound through movements see a decrease in queue with indicates this may be true. Semi actuated scenarios for the bicyclist configura tion show little improvement ( F igure 4 16). -20 0 20 40 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Semi actuated Pedestrian vs. Base LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI

PAGE 60

60 Figure 4 15 Base design vs. bicyclist design queue changes pre timed. Figure 4 16 Base design vs. bicyclist design queue changes semi actuated. The combination configuration has queue increases for almost every movement in pre timed settings except for the southbound through/right movement in heavy pedestrian and bicyclist scenarios ( F igure 4 17). There are no significant changes for -40 -20 0 20 40 60 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Pre timed Bicyclist vs. Base LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET -40 -20 0 20 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Semi actuated Bicyclist vs. Base LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI

PAGE 61

61 semi actuated scenarios ( Fi gure 4 18) except for a d ecrease in queue for northbound movements in the maximum volume setting. Figure 4 17 Base design vs. combination design queue changes pre timed. Figure 4 1 8 Base design vs. combination design queue changes semi actuated. The alternate configuration has an abnormal increase in queue for southbound through/right movements in the maximum volume scenario for a pre timed signal system ( Fi gure 4 19 ) The opposite effect occurs with medium vehicles The alternative -40 -20 0 20 40 60 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Pre timed Combination vs. Base LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET -40 -20 0 20 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Semi actuated Combination vs. Base LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI

PAGE 62

62 design has the same relative green time for the north and south movements to the other two phases. The simulation of the alternative design shows that the southbound queues increase because southbound right turning vehicles flow are impeded by concurrent b icyclist movements Once vehicles clear the bicyclist traffic, pedestrian traffic continues to impede flow. This makes vehicles back up further than in the base scenario, which leads to cycle failure Simulations of the HP HB HV PRET scenario without bicyclists present showed the SB T/R queue length increase fully dissipate. Further testing revealed that by increasing the green time for the phase by up to 15 seconds all cycle failure is removed without c ausing any considerable changes to other operations When comparing the alternative configuration to the combination configuration, ( both have the same offset crosswalk distance ) the combination con figuration decreases queueing. This is because the combination configuration includes bicyclists in the setback crossings. The semi actuated results for the alternative configuration show no changes in queueing in F igure 4 20 Figure 4 1 9 Base design vs. alternative design queue changes pre timed. -60 -40 -20 0 20 40 60 80 100 120 140 160 180 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Pre timed Alternative vs. Base LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET

PAGE 63

63 Figure 4 20 Base design vs. alternative design queue changes semi actuated. The queueing measure of effectiveness shows the bicyclist configuration does the best job in queue reduction for the s ystem in both pre timed and semi actuated scenarios. This is because vehicles can freely begin to move from a complete stop and yield to any bikes and pedestrians downstream. In the base configuration, vehicles would wait for the bikes to complete their th rough movement before moving. The pedestrian configurations queues mostly increase, similar to the combination configuration. The alternative design mostly stays the same as the base except for one outlier. In all systems and configurations, the most sensi tive mode and volume to geometric changes were high volumes of vehicles. Average Vehicle Speed The average speed of vehicles in a system is the last measure of effectiveness. Vehicle average speed is a good indicator of if vehicles benefited from any changes to a signal or geometry. Average vehicle speed is not typically used for analyzing isolated intersections but for this research was determined to be significant Although delay can -20 0 20 40 NBL/T/R SBL SBT/R EBL EBT/R WBL WBT WBR Change in Queue (feet) Movement Change in Queue Semi actuated Alternative vs. Base LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI

PAGE 64

64 be used to indirectly measure speed characteristics the average speed can also indicate if vehicles are seeing improved flow with minor geometric changes. The pedestrian scenario results shown in F igure 4 21 show vehicle average speed reducing in most scenario s This is ironic given the large reduction in vehicle delay observed This observation occurs because vehicles clear the stop bar and then slow down to turn past bicycles and then again for pedestrians. Figure 4 21 Vehicle average speed fluctuations base vs. pedestrian. The bicyclist scenarios have vehicle average speeds either staying the same or rising except for low pedestrian and bicyclist scenarios in pre timed signals ( Fi gure 4 22) and the semi actuated results show slight net increases Once volumes of pedestrians and bicyclists are increased, the average speed increases for vehicles. This may be because vehicles do have slightly more turning space and therefore do not impede through moving veh icles. 8 9 10 11 12 13 14 15 16 17 18 19 20 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Vehicle Average Speed (mph) Scenario Vehicle Average Speed Changes Pedestrian vs. Base Base Configuration Pedestrian Configuration

PAGE 65

65 Figure 4 22 Vehicle average speed fluctuations base vs. bicyclist. The combination configuration results in Fi gure 4 23 show varying changes in vehicle average speed. The pre timed results show mainly decreasing average speed. This decrease in pre timed scenarios is because vehicles must slow down while wai ting for bicyclists and pedestrians to clear, similar to the pedestrian intersection. Figure 4 23 Vehicle average speed fluctuations base vs. combination. 8 9 10 11 12 13 14 15 16 17 18 19 20 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Vehicle Average Speed (mph) Scenario Vehicle Average Speed Changes Bicyclist vs. Base Base Configuration Bicyclist Configuration 8 9 10 11 12 13 14 15 16 17 18 19 20 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Vehicle Average Speed (mph) Scenario Vehicle Average Speed Changes Combination vs. Base Base Configuration Combination Configuration

PAGE 66

66 The alternat ive results for average speed in F igure 4 24 show little to no improvements to any scenario. Figure 4 24 Vehicle average speed fluctuations base vs. alternative. The lack of improvement for vehicle average speed in the alternative design stem s from the queueing issues derived from bicyclist movements impeding use of the extra right turn queue storage. For vehicle average speeds, the best improvements come from semi actuated signals in all configurations Pre timed signals mostly showed decreases in vehicle average speed. Semi actuated scenarios are not as sensitive to vehicles that have to slow down for turns due to pedestrians or bicyclists beca use the queue reductions correlate to higher vehicle speeds. The delay comparisons show great improvements in the pedestrian configuration for pre timed signals. The best queue reductions come from bicyclist configurations. Stanek (2015) research found that motor vehicle delay could be reduced when right turn yields are implemented with downstream crossings for 8 9 10 11 12 13 14 15 16 17 18 19 20 LP LB MV PRET LP LB HV PRET HP HB MV PRET HP HB HV PRET LP LB MV SEMI LP LB HV SEMI HP HB MV SEMI HP HB HV SEMI Vehicle Average Speed (mph) Scenario Vehicle Average Speed Changes Alternative vs. Base Base Configuration Alternative Configuration

PAGE 67

67 bicyclists The configuration used by the researcher is similar to the combination configuration in this research. This research f ound that in the bicyclist configuration, queues can be reduced and v ehicle average speed can go up The combination configuration found weighted mode delay and total vehicle delay can drop for virtually all scenarios. Stanek and Alexander (2015) also reaf firm literature review that a sole phase for all pedestrian and bicyclist movements would make delay worsen for all modes. Semi actuated signals fair better than pre timed signals for most configurations with vehicle average speed increases. The results f or the alternative intersection were not expected. Figure 4 25 shows what happens during the VISSIM simulation in this configuration. Figure 4 25. Alternat ive intersection right turn queueing

PAGE 68

68 This design shown in Fi gure 4 25 utilizes a median refuge island with a staggered crosswalk. The white vehicle that made an eastbound right turn ( see red circle ) can stop after completing the turn while eastbound through moving vehicles continue unimpeded. The expected result wa s for the alternate configuration to mimic those of the pedestrian configuration. Raw data results from simulations are summarized in the appendix T ables A 1 to A 40 and screenshots of VISSIM configurations are provided in Fi gure A 1 to A 1 0.

PAGE 69

69 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS Chapter 5 focuses on interpretation of results. Recommendations are provided based on trends found in results. The information provided is meant to help engineers make more informed decision s when weighing how operati on s are affected by the intersection design. The results from the analyses indicate that some of the newer configurations perform better when compared to the base configuration. The primary goal of the analysis is to identify scenarios th at result in better operations under each particular type of design. One common issue in shared through/right lane s is when right turning vehicle s can not complete a movement because of concurrent pedestrian movements. This cause s b rief queue s of through moving vehicles behind the right turning vehicle as shown in F igure 5 1. Figure 5 1. Base configuration right turn issue. The eastbound moving black vehicle is attempting to turn right but is blocked by pedestrians. This results in the vehicles behind the black vehicle queueing until the black car can turn, or the queued vehicles can change lanes. Changes to the base

PAGE 70

70 configuration show encouraging results. Table 5 1 summarizes operationa l effects pros and cons from the different configurations followed by more in depth conclusions Table 5 1. Summarized results by configuration The pedestrian configuration aimed to give right turning v ehicles space to turn while letting bicyclists remain in line with vehicular traffic. The added benefit of this configuration is right turning cars face pedestrians directly This is safer for the pedestrians who can clear the cr ossing before a vehicle approaches S ight distance for queued vehicles shortens The pedestrian configuration will reduce total weighted delay for pre timed signals and see no significant changes for semi actuated signals. Total vehicle delay is r educed for a pre timed signal or semi actuated signal with this geometric adjustment. Vehicle queues increase for critical volume movements, Configuration Description Pros Cons Pedestrian Reduction in combined weighted delay for pre timed signals and no change to semi actuated signals. Reduction in total vehicle delay in all scenarios. Increase in vehicle queueing for critical volume movements in pre timed signals. Reduction in vehicle average speed for all scenarios. Bicyclist Reduct ion in vehicle queueing for most pre timed and semi actuated scenarios. Vehicle average speed increases for all scenarios except for LP LB pre timed scenarios. C ombined weighted delay increase for semi actuated scenarios. Total vehicle dela y increase for high vehicle volumes in semi actuated scenarios. Minor street through movements see large queue increase in LP LB HV PRET. Combination Reduction in combined weighted delay for all scenarios except for HP HB HV SEMI. Reduction in total vehicle delay for all scenarios. I nc rease in vehicle average speed for most semi actuated scenarios Vehicle queue increases for most pre timed scenarios m ovements. Vehicle average speed reduction in high vehicle volumes in pre timed scenarios. Alternative Slight reduction in total vehicle delay for LP LB MV SEMI scenario. Slight increase in vehicle average speed for HP HB MV SEMI scenario. Increase in combined weighted delay for most scenarios. Total vehicle delay increases for all scenarios. Slight increase in vehicle queueing for pre timed scenarios Vehicle average speed reduc es in some scenarios.

PAGE 71

71 particularly for the heaviest volume scenario for a pre t imed signal and no significant changes for semi actuated signals. Vehicle average speed will drop for pre timed and semi actuated scenarios due to the reduction in distance between nodes. In the bicyclist configuration, bicycles cross downstream and in line with pedestrians. Th is design reduce s right tu rning conflicts with vehicles and bikes. The shorter crossing distance for bicyclists is safer because of the separation from vehicles No quantifiable safety measurements are recorded however because this research focuses on operational effects. There is no significant change in total weighted delay for this configuration due to little geometric changes made. No significant changes in total vehicle delay exist except for high vehicle volume semi actuated scenarios where delay rises. This is becau se the increase of turning movements creates more conflicts with bikes at the crosswalk that normally would not occur. This benefits queueing of vehicles because vehicles can clear the intersection immediately. Queues reduce or stay the same for all semi a ctuated and pre timed scenarios except for the minor street through/right movements in a LP LB HV pre timed setting. This particular scenario has high increases because through moving vehicles in shared lanes are impeded by right turning movements. The sha red lane for the eastbound direction sees no increase because of a relatively low right turn to through movement ratio compared to minor street movements. Vehicle average speed mostly increases for all scenarios due to the reduction in queueing. The combin ation configuration utilizes pedestrian an d bicyclist friendly design This design was tested to see if benefits gained from both of the previous designs would hold when together Th e design force s bicyclists to travel further relative distances This

PAGE 72

72 conf iguration experiences drops in total weighted delay and total vehicle delay for all scenarios as expected. Queues do not change for semi actuated scenarios but do rise slightly for pre timed scenarios with high vehicle volumes, similar to the pedestrian co nfiguration but to a lesser degree This occurs because vehicles have more space to clear the stop bar and complete a turn versus the pedestrian configuration where bicyclists can impede flow more. The alternative configuration was designed keep benefits of the pedestrian intersection but let vehicles have similar sight distance as the base configuration. This configuration forces pedestrians to travel in a staggered crossing The only two main differences from the alternative design to the pedestrian des ign is that vehicles do not have a setback stop bar and pedestrians must travel a further distance. Individual vehicle delay does not change significantly compared to the base scenario. This configuration increases total weighted delay for all scenarios an d slightly increases total vehicle delay for some scenarios. No vehicle queue changes occur for semi actuated scenarios or pre timed scenarios with the exception of one outlier for the heaviest volume category for southbound through/right vehicle movements This particular scenario movement crosses a tipping point where vehicles that turn right have little to no gap opportunities to complete their respective turn. This causes cycle failure. This configuration will increase vehicle queues for shared through/ right lanes if there is a large proportion of right turning vehicles with heavy pedestrian and bicyclist movement. A benefit and cost analysis is not included in this research but the cost of right of way should be considered when comparing how intrusive t he designs are to any surrounding property

PAGE 73

73 T he following recommendations of this research are : The pedestrian configuration will reduce total vehicle delay in all scenarios and reduce combined weighted delay for all modes in pre timed signal systems The configuration will also increase queueing for high vehicle volume scenarios and slightly reduce sight distance for vehicles The bicyclist configuration will reduce vehicle queueing in most scenarios, regardless of the signal system. Vehicle average speed will also increase. Delay for vehicles will increase in heavy vehicle, semi actuated scenarios and combined weighted delay will increase slightly for most scenarios. The combination configuration will reduce vehicle delay for all scenarios but increase bicyclist delay for all scenarios. This configuration involves added distance for movements. Vehicle queues are not affected in semi actuated settings, and generally rise in pre timed settings. The alternative configuration offers no beneficial operational effect to an intersection for any mode. Queue increases are minimal but dissipate if no and changes to signage to not confuse pedestrians. Future research can find the ideal volume scenarios or ideal geometric changes where certain operational effects are maximized. Adaptive signal control technology can be simulated in the future to see h ow that technologies efficiency is effected by these configurations. A technology that allows crossing times for pedestrians to represent the amount present and not historically expecte d can potentially have a significant impact to operational effects for the system as well. All results are simulated in VISSIM and should be tested in a real world environment to confirm results

PAGE 74

74 APPENDIX RESULT SHEETS Table A 1. Base pre timed, medium veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 15.79 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 10.35 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 30.16 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 17.69 Geometric Layout Base Cycle Length (s) 85 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 24.5 19.53 *NB East 36.57 NB 19.73 NBT 24.5 11.85 NB West 38.53 SB 19.14 NBR 24.5 11.84 SB East 37.13 EB 17.10 SBL 3.4 21.68 SB West 36.84 WB 15.30 SBT 55.4 6.45 EB North 25.95 SBR 55.4 6.78 EB South 25.02 EBL 20.4 29.26 WB North 27.06 EBT 49.1 8.42 WB South 25.67 EBR 49.1 8.66 WBL 10.0 29.28 WBT 30.3 8.02 WBR 2.1 15.17 *NB East denotes NB travel on the eastern most crosswalk

PAGE 75

75 Table A 2. Base, pre timed, medium veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 14.35 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 11.09 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 29.56 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 18.11 Geometric Layout Base Cycle Length (s) 85 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 28.6 25.57 *NB East 36.91 NB 20.12 NBT 28.6 13.17 NB West 37.34 SB 20.53 NBR 28.6 13.22 SB East 37.14 EB 16.58 SBL 4.1 28.35 SB West 37.50 WB 16.03 SBT 95.3 6.69 EB North 25.48 SBR 95.3 10.52 EB South 25.67 EBL 20.2 30.22 WB North 25.53 EBT 50.0 8.66 WB South 26.14 EBR 50.0 8.96 WBL 10.4 28.04 WBT 31.0 8.49 WBR 1.8 19.04 *NB East denotes NB travel on the eastern most crosswalk

PAGE 76

76 Table A 3. Base, pre timed, high veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 12.54 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 9.66 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 35.40 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 22.63 Geometric Layout Base Cycle Length (s) 120 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 79.0 26.72 *NB East 52.23 NB 28.30 NBT 79.0 14.33 NB West 51.50 SB 29.83 NBR 79.0 12.91 SB East 52.32 EB 18.34 SBL 6.8 27.47 SB West 50.74 WB 16.29 SBT 108.0 6.00 EB North 26.58 SBR 108.0 7.88 EB South 24.85 EBL 66.4 31.71 WB North 25.76 EBT 114.3 6.12 WB South 27.06 EBR 114.3 4.74 WBL 21.5 33.97 WBT 53.7 6.34 WBR 3.3 16.63 *NB East denotes NB travel on the eastern most crosswalk

PAGE 77

77 Table A 4. Base, pre timed, high veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 11.54 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 10.11 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 34.19 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 23.50 Geometric Layout Base Cycle Length (s) 120 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 112.2 29.93 *NB East 48.25 NB 30.21 NBT 112.2 14.35 NB West 50.37 SB 29.02 NBR 112.2 16.10 SB East 50.24 EB 18.57 SBL 8.3 33.37 SB West 48.27 WB 18.64 SBT 141.1 6.59 EB North 26.70 SBR 141.1 9.47 EB South 27.28 EBL 71.4 30.89 WB North 26.31 EBT 116.5 6.28 WB South 26.23 EBR 116.5 7.39 WBL 20.5 34.83 WBT 55.9 6.23 WBR 3.2 21.44 *NB East denotes NB travel on the eastern most crosswalk

PAGE 78

78 Tab le A 5. Base, semi actuated, medium veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 19.19 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 8.54 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 25.34 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 12.01 Geometric Layout Base Cycle Length (Critical) (s) 57 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 12.1 14.72 *NB East 31.01 NB 12.49 NBT 12.1 8.55 NB West 33.80 SB 12.42 NBR 12.1 8.25 SB East 30.52 EB 12.34 SBL 2.2 17.02 SB West 29.60 WB 11.01 SBT 31.5 6.95 EB North 23.62 SBR 31.5 6.21 EB South 21.78 EBL 6.7 17.53 WB North 22.12 EBT 32.1 7.48 WB South 22.40 EBR 32.1 7.48 WBL 3.7 23.44 WBT 19.6 6.56 WBR 1.3 11.21 *NB East denotes NB travel on the eastern most crosswalk

PAGE 79

79 Table A 6. Base, semi actuated, medium veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 17.98 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 9.53 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 25.55 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 12.21 Geometric Layout Base Cycle Length (Critical) (s) 57 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 13.8 22.05 *NB East 31.01 NB 12.01 NBT 13.8 9.18 NB West 31.60 SB 12.37 NBR 13.8 9.75 SB East 31.11 EB 11.90 SBL 2.2 19.17 SB West 32.10 WB 12.54 SBT 37.5 6.98 EB North 22.64 SBR 37.5 10.42 EB South 22.91 EBL 9.7 20.73 WB North 22.15 EBT 36.4 7.94 WB South 22.47 EBR 36.4 11.54 WBL 5.5 22.39 WBT 20.7 7.18 WBR 1.5 18.09 *NB East denotes NB travel on the eastern most crosswalk

PAGE 80

80 Table A 7. Base, semi actuated, high veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 13.05 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 7.66 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 43.54 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 28.22 Geometric Layout Base Cycle Length (Critical) (s) 192 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 112.9 32.09 *NB East 83.24 NB 46.16 NBT 112.9 14.86 NB West 93.52 SB 44.84 NBR 112.9 14.44 SB East 86.17 EB 13.68 SBL 6.4 34.88 SB West 94.91 WB 14.32 SBT 163.5 8.82 EB North 21.65 SBR 163.5 8.56 EB South 20.85 EBL 10.6 12.93 WB North 21.54 EBT 91.7 4.43 WB South 18.63 EBR 91.7 3.88 WBL 9.2 21.05 WBT 47.7 4.46 WBR 2.4 11.74 *NB East denotes NB travel on the eastern most crosswalk

PAGE 81

81 Table A 8. Base, semi actuated, high veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 11.50 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.58 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 46.67 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 27.96 Geometric Layout Base Cycle Length (Critical) (s) 192 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 163.5 36.86 *NB East 93.63 NB 40.63 NBT 163.5 16.49 NB West 93.68 SB 41.49 NBR 163.5 14.00 SB East 95.31 EB 17.77 SBL 7.6 35.60 SB West 92.11 WB 17.06 SBT 179.6 8.62 EB North 23.32 SBR 179.6 11.09 EB South 22.88 EBL 13.4 16.49 WB North 23.07 EBT 109.6 4.85 WB South 23.91 EBR 109.6 4.83 WBL 9.9 25.74 WBT 57.0 4.98 WBR 3.0 17.12 *NB East denotes NB travel on the eastern most crosswalk

PAGE 82

82 Figure A 1. Base configuration on VISSIM as pre timed. Figure A 2. Base configuration on VISSIM as semi actuated.

PAGE 83

83 Table A 9. Pedestrian, pre timed, medium veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 15.01 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 7.48 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 31.12 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 15.05 Geometric Layout Pedestrian Cycle Length (s) 90 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 26.1 24.78 *NB East 41.13 NB 15.63 NBT 26.1 11.27 NB West 41.80 SB 12.26 NBR 26.1 9.83 SB East 36.58 EB 17.61 SBL 3.8 18.83 SB West 39.65 WB 14.20 SBT 57.4 4.65 EB North 25.87 SBR 57.4 6.17 EB South 25.30 EBL 29.0 20.02 WB North 26.86 EBT 56.0 4.97 WB South 26.29 EBR 56.0 4.60 WBL 11.1 24.65 WBT 34.0 5.75 WBR 2.4 14.16 *NB East denotes NB travel on the eastern most crosswalk

PAGE 84

84 Table A 10. Pedestrian, pre timed, medium veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 13.98 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 10.13 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 30.00 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 17.93 Geometric Layout Pedestrian Cycle Length (s) 90 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 30.1 26.85 *NB East 37.56 NB 19.30 NBT 30.1 11.96 NB West 39.79 SB 18.71 NBR 30.1 14.28 SB East 38.65 EB 17.36 SBL 4.6 24.90 SB West 38.88 WB 16.78 SBT 93.4 6.51 EB North 25.08 SBR 93.4 11.66 EB South 26.23 EBL 22.9 29.82 WB North 25.77 EBT 56.2 8.10 WB South 25.26 EBR 56.2 8.21 WBL 11.0 23.97 WBT 31.9 6.12 WBR 1.9 16.73 *NB East denotes NB travel on the eastern most crosswalk

PAGE 85

85 Table A 11. Pedestrian, pre timed, high veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 11.84 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.48 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 35.04 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 23.86 Geometric Layout Pedestrian Cycle Length (s) 130 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 82.3 27.32 *NB East 45.53 NB 29.21 NBT 82.3 15.28 NB West 49.09 SB 29.00 NBR 82.3 10.74 SB East 48.33 EB 20.16 SBL 5.9 21.28 SB West 47.81 WB 19.39 SBT 110.7 5.43 EB North 26.46 SBR 110.7 7.69 EB South 26.05 EBL 92.1 29.17 WB North 29.20 EBT 132.0 5.66 WB South 29.48 EBR 132.0 4.58 WBL 27.1 25.97 WBT 65.7 4.46 WBR 2.9 12.92 *NB East denotes NB travel on the eastern most crosswalk

PAGE 86

86 Table A 12. Pedestrian pre timed, high veh high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 10.17 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 9.14 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 34.36 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 22.61 Geometric Layout Pedestrian Cycle Length (s) 130 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 113.1 33.10 *NB East 47.40 NB 27.41 NBT 113.1 14.98 NB West 48.40 SB 24.16 NBR 113.1 14.62 SB East 49.47 EB 20.87 SBL 6.3 30.75 SB West 47.21 WB 19.35 SBT 217.6 6.69 EB North 26.97 SBR 217.6 9.39 EB South 28.20 EBL 108.1 29.77 WB North 28.14 EBT 147.7 5.87 WB South 27.48 EBR 147.7 4.53 WBL 24.6 26.43 WBT 64.3 4.49 WBR 3.3 18.06 *NB East denotes NB travel on the eastern most crosswalk

PAGE 87

87 Table A 13. Pedestrian, semi actuated, medium veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 18.98 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 7.99 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 25.34 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 12.79 Geometric Layout Pedestrian Cycle Length (Critical) (s) 61 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 14.6 16.58 *NB East 31.55 NB 12.60 NBT 14.6 8.44 NB West 34.97 SB 13.52 NBR 14.6 9.25 SB East 33.02 EB 12.74 SBL 2.3 16.08 SB West 34.04 WB 12.41 SBT 35.6 5.36 EB North 20.57 SBR 35.6 6.00 EB South 21.64 EBL 5.9 16.66 WB North 22.46 EBT 32.8 7.42 WB South 20.60 EBR 32.8 7.12 WBL 3.5 20.49 WBT 19.1 5.74 WBR 1.2 9.77 *NB East denotes NB travel on the eastern most crosswalk

PAGE 88

88 Table A 14. Pedestrian, semi actuated, medium veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 18.12 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 8.91 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 25.80 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 13.21 Geometric Layout Pedestrian Cycle Length (Critical) (s) 61 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 14.0 18.11 *NB East 31.50 NB 12.73 NBT 14.0 8.96 NB West 32.76 SB 14.10 NBR 14.0 10.06 SB East 33.73 EB 13.35 SBL 2.7 18.65 SB West 32.87 WB 12.69 SBT 39.8 5.80 EB North 23.01 SBR 39.8 8.67 EB South 22.58 EBL 6.1 19.13 WB North 22.23 EBT 34.1 7.70 WB South 21.42 EBR 34.1 10.81 WBL 5.2 24.94 WBT 19.9 6.07 WBR 1.3 14.82 *NB East denotes NB travel on the eastern most crosswalk

PAGE 89

89 Table A 15. Pedestrian, semi actuated, high veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 12.46 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 7.29 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 48.13 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 28.75 Geometric Layout Pedestrian Cycle Length (Critical) (s) 207 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 118.3 32.11 *NB East 103.04 NB 43.81 NBT 118.3 14.88 NB West 97.61 SB 47.65 NBR 118.3 11.97 SB East 104.58 EB 14.11 SBL 6.7 27.35 SB West 100.66 WB 14.80 SBT 179.2 6.28 EB North 20.22 SBR 179.2 6.90 EB South 19.84 EBL 9.2 14.05 WB North 20.62 EBT 99.5 4.43 WB South 19.87 EBR 99.5 5.43 WBL 8.5 28.28 WBT 53.1 3.73 WBR 2.5 10.43 *NB East denotes NB travel on the eastern most crosswalk

PAGE 90

90 Table A 16. Pedestrian, semi actuated, high veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 11.07 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.39 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 48.76 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 27.38 Geometric Layout Pedestrian Cycle Length (Critical) (s) 207 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 173.2 37.52 *NB East 99.62 NB 36.97 NBT 173.2 15.18 NB West 101.54 SB 41.42 NBR 173.2 15.84 SB East 101.76 EB 17.72 SBL 8.7 28.47 SB West 101.65 WB 17.82 SBT 201.0 6.57 EB North 23.17 SBR 201.0 9.05 EB South 21.92 EBL 11.0 18.70 WB North 21.55 EBT 106.6 4.70 WB South 22.71 EBR 106.6 7.59 WBL 11.4 35.48 WBT 56.3 3.93 WBR 2.8 17.64 *NB East denotes NB travel on the eastern most crosswalk

PAGE 91

91 Figure A 3 Pedestrian configuration on VISSIM as pre timed. Figure A 4. Pedestrian configuration on VISSIM as semi actuated

PAGE 92

92 Table A 17. Bicyclist, pre timed, medium veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 15.74 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 10.45 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 30.02 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 17.31 Geometric Layout Bicyclist Cycle Length (s) 85 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 23.5 19.38 *NB East 38.37 NB 19.21 NBT 23.5 11.44 NB West 38.13 SB 19.13 NBR 23.5 11.78 SB East 39.67 EB 15.86 SBL 3.8 21.99 SB West 36.04 WB 15.64 SBT 54.4 6.22 EB North 25.90 SBR 54.4 7.02 EB South 25.64 EBL 21.8 29.98 WB North 27.46 EBT 50.0 8.55 WB South 24.85 EBR 50.0 6.59 WBL 10.8 30.83 WBT 29.9 8.02 WBR 2.2 15.16 *NB East denotes NB travel on the eastern most crosswalk

PAGE 93

93 Table A 18. Bicyclist, pre timed, medium veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 15.17 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 10.84 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 29.54 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 18.25 Geometric Layout Bicyclist Cycle Length (s) 85 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 25.8 21.38 *NB East 36.70 NB 20.22 NBT 25.8 12.09 NB West 36.77 SB 19.49 NBR 25.8 14.15 SB East 36.81 EB 17.14 SBL 3.6 24.64 SB West 36.70 WB 16.70 SBT 67.8 6.93 EB North 25.72 SBR 67.8 8.25 EB South 25.68 EBL 20.3 29.08 WB North 26.52 EBT 50.0 8.68 WB South 25.62 EBR 50.0 11.00 WBL 10.0 29.35 WBT 30.3 8.03 WBR 2.1 20.65 *NB East denotes NB travel on the eastern most crosswalk

PAGE 94

94 Table A 19. Bicyclist, pre timed, high veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 12.24 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 9.49 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 35.98 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 25.18 Geometric Layout Bicyclist Cycle Length (s) 120 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 108.6 25.79 *NB East 49.91 NB 37.32 NBT 108.6 15.34 NB West 55.68 SB 36.31 NBR 108.6 15.14 SB East 56.73 EB 15.16 SBL 8.2 29.69 SB West 56.14 WB 16.45 SBT 155.3 7.07 EB North 24.93 SBR 155.3 7.72 EB South 27.45 EBL 56.6 30.11 WB North 27.00 EBT 92.7 5.69 WB South 25.47 EBR 92.7 4.50 WBL 20.2 33.95 WBT 47.3 5.91 WBR 2.5 12.63 *NB East denotes NB travel on the eastern most crosswalk

PAGE 95

95 Table A 20. Bicyclist pre timed, high veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 12.03 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 10.03 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 33.94 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 24.16 Geometric Layout Bicyclist Cycle Length (s) 120 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 82.5 24.61 *NB East 48.88 NB 30.33 NBT 82.5 14.54 NB West 51.28 SB 31.41 NBR 82.5 17.04 SB East 48.01 EB 18.66 SBL 7.4 31.70 SB West 50.31 WB 19.02 SBT 139.1 6.47 EB North 26.27 SBR 139.1 8.52 EB South 25.90 EBL 62.4 31.08 WB North 25.94 EBT 107.8 6.36 WB South 26.48 EBR 107.8 6.69 WBL 22.3 33.10 WBT 58.8 6.37 WBR 3.1 20.44 *NB East denotes NB travel on the eastern most crosswalk

PAGE 96

96 Table A 21. Bicyclist, semi actuated, medium veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 19.35 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 8.60 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 27.05 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 12.42 Geometric Layout Bicyclist Cycle Length (Critical) (s) 57 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 14.8 17.27 *NB East 46.88 NB 13.55 NBT 14.8 7.34 NB West 35.44 SB 14.09 NBR 14.8 8.13 SB East 37.40 EB 11.80 SBL 2.3 17.68 SB West 32.32 WB 10.84 SBT 31.4 8.34 EB North 21.87 SBR 31.4 8.56 EB South 22.02 EBL 6.9 15.54 WB North 21.69 EBT 31.0 6.98 WB South 20.42 EBR 31.0 6.62 WBL 4.2 20.38 WBT 18.7 7.90 WBR 1.2 10.22 *NB East denotes NB travel on the eastern most crosswalk

PAGE 97

97 Table A 22. Bicyclist, semi actuated, medium veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 18.14 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 9.68 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 25.41 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 12.97 Geometric Layout Bicyclist Cycle Length (Critical) (s) 57 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 14.0 16.08 *NB East 31.39 NB 12.85 NBT 14.0 8.47 NB West 31.57 SB 13.07 NBR 14.0 7.63 SB East 30.80 EB 13.03 SBL 2.1 19.05 SB West 31.15 WB 12.92 SBT 33.6 8.21 EB North 23.19 SBR 33.6 10.19 EB South 22.33 EBL 8.5 20.48 WB North 22.00 EBT 37.0 7.75 WB South 22.25 EBR 37.0 10.55 WBL 6.9 28.36 WBT 19.9 8.38 WBR 1.7 16.33 *NB East denotes NB travel on the eastern most crosswalk

PAGE 98

98 Table A 23. Bicyclist, semi actuated, high veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 13.32 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.40 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 43.57 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 27.72 Geometric Layout Bicyclist Cycle Length (Critical) (s) 192 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 104.7 32.41 *NB East 88.47 NB 45.92 NBT 104.7 14.71 NB West 92.60 SB 43.92 NBR 104.7 13.59 SB East 96.39 EB 14.08 SBL 6.4 34.84 SB West 85.71 WB 13.69 SBT 165.0 10.77 EB North 20.34 SBR 165.0 12.70 EB South 19.20 EBL 10.2 14.17 WB North 21.13 EBT 84.3 4.18 WB South 21.48 EBR 84.3 3.81 WBL 8.3 32.16 WBT 44.6 5.30 WBR 2.1 11.91 *NB East denotes NB travel on the eastern most crosswalk

PAGE 99

99 Table A 24. Bicyclist, semi actuated, high veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 11.90 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 9.57 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 45.21 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 28.30 Geometric Layout Bicyclist Cycle Length (Critical) (s) 192 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 143.7 32.29 *NB East 91.57 NB 43.39 NBT 143.7 16.66 NB West 93.80 SB 45.23 NBR 143.7 14.87 SB East 91.52 EB 15.24 SBL 7.9 39.29 SB West 90.42 WB 15.99 SBT 173.2 11.14 EB North 21.48 SBR 173.2 12.82 EB South 22.33 EBL 11.7 20.01 WB North 22.32 EBT 96.2 4.42 WB South 21.10 EBR 96.2 7.34 WBL 14.1 42.44 WBT 52.8 5.57 WBR 2.7 18.33 *NB East denotes NB travel on the eastern most crosswalk

PAGE 100

100 Figure A 5 Bicyclist configuration on vissim as pre timed. Figure A 6. Bicyclist configuration on vissim as semi actuated

PAGE 101

101 Table A 25. Combination pre timed, medium veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 15.14 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 9.45 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 30.96 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 18.90 Geometric Layout Combination Cycle Length (s) 90 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 26.9 24.04 *NB East 41.70 NB 19.64 NBT 26.9 10.70 NB West 41.65 SB 20.02 NBR 26.9 9.67 SB East 36.63 EB 18.78 SBL 3.7 21.36 SB West 39.49 WB 17.53 SBT 54.9 5.53 EB North 27.11 SBR 54.9 7.15 EB South 25.01 EBL 28.8 30.14 WB North 27.43 EBT 56.1 7.73 WB South 26.45 EBR 56.1 7.26 WBL 11.1 25.04 WBT 33.9 6.01 WBR 2.4 15.53 *NB East denotes NB travel on the eastern most crosswalk

PAGE 102

102 Table A 26. Combination, pre timed, medium veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 14.88 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 9.91 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 30.04 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 19.59 Geometric Layout Combination Cycle Length (s) 90 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 33.7 26.74 *NB East 38.04 NB 22.07 NBT 33.7 11.24 NB West 38.81 SB 21.89 NBR 33.7 12.39 SB East 38.79 EB 17.90 SBL 3.2 22.17 SB West 40.40 WB 17.48 SBT 60.2 6.63 EB North 25.14 SBR 60.2 7.89 EB South 25.93 EBL 23.7 30.13 WB North 26.20 EBT 53.8 7.94 WB South 24.87 EBR 53.8 9.05 WBL 11.2 24.50 WBT 33.3 6.05 WBR 2.1 18.92 *NB East denotes NB travel on the eastern most crosswalk

PAGE 103

103 Table A 27. Combination, pre timed, high veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 11.62 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.57 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 33.45 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 24.97 Geometric Layout Combination Cycle Length (s) 130 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 88.1 26.46 *NB East 45.33 NB 32.52 NBT 88.1 12.28 NB West 47.35 SB 31.95 NBR 88.1 10.81 SB East 43.28 EB 20.01 SBL 6.8 26.52 SB West 48.59 WB 18.58 SBT 109.4 5.85 EB North 27.52 SBR 109.4 6.53 EB South 26.89 EBL 105.6 29.88 WB North 28.35 EBT 140.2 5.69 WB South 26.20 EBR 140.2 4.95 WBL 27.4 25.65 WBT 66.1 4.57 WBR 3.4 13.85 *NB East denotes NB travel on the eastern most crosswalk

PAGE 104

104 Table A 28. Combination, pre timed, high veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 10.75 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.95 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 34.16 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 25.69 Geometric Layout Combination Cycle Length (s) 130 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 153.8 31.05 *NB East 49.49 NB 32.97 NBT 153.8 12.81 NB West 48.25 SB 31.64 NBR 153.8 16.02 SB East 46.29 EB 20.32 SBL 5.7 26.87 SB West 47.52 WB 20.78 SBT 114.8 6.02 EB North 28.47 SBR 114.8 9.26 EB South 27.76 EBL 106.4 29.46 WB North 26.50 EBT 141.4 5.79 WB South 27.05 EBR 141.4 7.00 WBL 24.5 25.94 WBT 67.3 4.51 WBR 3.4 19.26 *NB East denotes NB travel on the eastern most crosswalk

PAGE 105

105 Table A 29. Combination, semi actuated, medium veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 19.15 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 7.91 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 24.69 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 11.86 Geometric Layout Combination Cycle Length (Critical) (s) 61 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 14.0 15.34 *NB East 32.92 NB 13.15 NBT 14.0 8.44 NB West 34.06 SB 13.20 NBR 14.0 7.83 SB East 31.55 EB 9.94 SBL 2.3 15.23 SB West 31.50 WB 11.71 SBT 32.5 5.21 EB North 20.73 SBR 32.5 5.61 EB South 20.80 EBL 6.3 16.92 WB North 21.09 EBT 32.6 7.42 WB South 20.75 EBR 32.6 8.40 WBL 3.6 19.90 WBT 19.5 5.72 WBR 1.3 10.24 *NB East denotes NB travel on the eastern most crosswalk

PAGE 106

106 Table A 30. Combination, semi actuated, medium veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 18.62 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 8.48 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 25.09 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 13.37 Geometric Layout Combination Cycle Length (Critical) (s) 61 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 16.1 17.85 *NB East 32.65 NB 14.39 NBT 16.1 8.98 NB West 32.96 SB 14.60 NBR 16.1 8.45 SB East 32.71 EB 12.04 SBL 1.9 15.78 SB West 33.23 WB 12.88 SBT 34.8 5.32 EB North 20.74 SBR 34.8 7.52 EB South 21.85 EBL 5.8 18.93 WB North 21.14 EBT 33.5 7.63 WB South 21.31 EBR 33.5 9.73 WBL 4.2 24.27 WBT 20.9 6.00 WBR 1.2 12.91 *NB East denotes NB travel on the eastern most crosswalk

PAGE 107

107 Table A 31. Combination, semi actuated, high veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 12.63 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 7.22 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 45.87 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 29.30 Geometric Layout Combination Cycle Length (Critical) (s) 207 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 125.8 40.29 *NB East 97.60 NB 50.42 NBT 125.8 14.06 NB West 97.71 SB 46.89 NBR 125.8 12.24 SB East 100.59 EB 13.88 SBL 7.3 27.52 SB West 100.27 WB 14.43 SBT 180.1 6.51 EB North 18.89 SBR 180.1 5.61 EB South 19.36 EBL 9.3 14.81 WB North 19.46 EBT 94.8 4.29 WB South 19.40 EBR 94.8 4.55 WBL 9.6 29.51 WBT 47.6 3.70 WBR 2.5 11.81 *NB East denotes NB travel on the eastern most crosswalk

PAGE 108

108 Table A 32. Combination, semi actuated, high veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 12.08 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 7.86 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 47.31 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 31.45 Geometric Layout Combination Cycle Length (Critical) (s) 207 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 128.7 35.49 *NB East 97.87 NB 48.58 NBT 128.7 15.07 NB West 97.09 SB 50.78 NBR 128.7 11.02 SB East 96.89 EB 16.08 SBL 6.8 26.36 SB West 98.19 WB 16.66 SBT 183.7 5.89 EB North 22.17 SBR 183.7 9.11 EB South 21.12 EBL 10.1 18.58 WB North 21.24 EBT 105.1 4.49 WB South 21.40 EBR 105.1 9.14 WBL 10.6 34.59 WBT 54.9 3.71 WBR 2.1 13.81 *NB East denotes NB travel on the eastern most crosswalk

PAGE 109

109 Figure A 7 Combination configuration on VISSIM as pre timed. Figure A 8. Combination configuration on VISSIM as semi actuated

PAGE 110

110 Table A 33. Alternative, pre timed, medium veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 14.93 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 10.87 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 35.91 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 19.56 Geometric Layout Alternate Cycle Length (s) 85 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 21.2 21.52 *NB East 39.38 NB 18.95 NBT 21.2 10.40 NB West 39.85 SB 18.00 NBR 21.2 10.90 SB East 39.79 EB 20.30 SBL 3.2 19.30 SB West 41.41 WB 20.53 SBT 46.9 5.52 EB North 35.01 SBR 46.9 6.56 EB South 34.53 EBL 28.3 33.42 WB North 33.82 EBT 63.5 9.11 WB South 31.80 EBR 63.5 7.67 WBL 11.1 31.77 WBT 40.7 9.15 WBR 2.3 16.73 *NB East denotes NB travel on the eastern most crosswalk

PAGE 111

111 Table A 34. Alternative, pre timed, medium veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 14.62 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 11.27 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 36.70 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 19.28 Geometric Layout Alternate Cycle Length (s) 85 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 23.4 24.28 *NB East 40.63 NB 18.36 NBT 23.4 11.25 NB West 40.66 SB 18.34 NBR 23.4 12.58 SB East 42.20 EB 20.37 SBL 3.0 22.84 SB West 40.84 WB 19.67 SBT 54.8 5.84 EB North 34.62 SBR 54.8 9.61 EB South 34.51 EBL 24.0 32.37 WB North 34.27 EBT 62.5 9.24 WB South 34.61 EBR 62.5 14.47 WBL 11.2 32.24 WBT 38.4 8.69 WBR 2.2 19.75 *NB East denotes NB travel on the eastern most crosswalk

PAGE 112

112 Table A 35. Alternative pre timed, high veh, low bike, low ped. Signal System Type Pre timed Vehicle Average Speed (mph) 12.04 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 9.67 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 37.61 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 23.18 Geometric Layout Alternate Cycle Length (s) 120 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 110.3 31.01 *NB East 54.05 NB 29.09 NBT 110.3 13.34 NB West 49.98 SB 30.34 NBR 110.3 14.87 SB East 55.67 EB 17.73 SBL 6.3 28.33 SB West 54.94 WB 18.10 SBT 122.3 6.20 EB North 28.31 SBR 122.3 8.18 EB South 29.48 EBL 64.2 31.44 WB North 28.99 EBT 109.7 6.08 WB South 31.13 EBR 109.7 4.57 WBL 20.4 35.66 WBT 56.9 6.41 WBR 2.9 14.18 *NB East denotes NB travel on the eastern most crosswalk

PAGE 113

113 Table A 36. Alternative pre timed, high veh, high bike, high ped. Signal System Type Pre timed Vehicle Average Speed (mph) 10.04 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 10.31 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 37.73 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 23.33 Geometric Layout Alternate Cycle Length (s) 120 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 107.0 34.61 *NB East 53.08 NB 29.28 NBT 107.0 15.71 NB West 54.31 SB 29.67 NBR 107.0 18.40 SB East 54.32 EB 18.72 SBL 7.3 30.91 SB West 52.90 WB 18.21 SBT 303.2 6.88 EB North 30.00 SBR 303.2 11.52 EB South 30.51 EBL 82.1 30.51 WB North 29.19 EBT 117.1 6.16 WB South 29.12 EBR 117.1 9.13 WBL 20.3 36.34 WBT 55.2 6.32 WBR 3.1 17.51 *NB East denotes NB travel on the eastern most crosswalk

PAGE 114

114 Table A 37. Alternative, semi actuated, medium veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 19.33 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 8.36 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 25.84 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 11.65 Geometric Layout Alternate Cycle Length (Critical) (s) 57 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 12.8 16.31 *NB East 41.17 NB 12.18 NBT 12.8 8.79 NB West 35.00 SB 11.65 NBR 12.8 7.80 SB East 31.46 EB 12.09 SBL 2.1 17.53 SB West 32.30 WB 10.71 SBT 30.2 6.93 EB North 20.51 SBR 30.2 7.69 EB South 21.98 EBL 6.9 15.85 WB North 20.50 EBT 32.1 7.35 WB South 21.80 EBR 32.1 6.80 WBL 3.9 19.50 WBT 19.5 6.57 WBR 1.2 11.44 *NB East denotes NB travel on the eastern most crosswalk

PAGE 115

115 Table A 38. Alternative, semi actuated, medium veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 17.10 Vehicle Volume Category Medium Weighted Veh Average Delay (s/veh) 10.19 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 31.98 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 13.77 Geometric Layout Alternate Cycle Length (Critical) (s) 57 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 12.4 18.91 *NB East 34.34 NB 11.38 NBT 12.4 8.00 NB West 33.59 SB 11.22 NBR 12.4 9.05 SB East 34.23 EB 15.79 SBL 2.0 20.59 SB West 33.53 WB 15.70 SBT 28.5 6.17 EB North 31.00 SBR 28.5 8.85 EB South 31.49 EBL 13.2 25.68 WB North 30.42 EBT 48.1 8.81 WB South 31.13 EBR 48.1 13.66 WBL 6.6 29.37 WBT 28.3 7.87 WBR 1.9 17.99 *NB East denotes NB travel on the eastern most crosswalk

PAGE 116

116 Table A 39. Alternative semi actuated, high veh, low bike, low ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 13.09 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 7.63 Pedestrian Volume Category Low Weighted Ped Average Delay (s/ped) 45.44 Bicyclist Volume Category Low Weighted Bike Average Delay (s/bike) 27.09 Geometric Layout Alternate Cycle Length (Critical) (s) 192 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 104.1 35.66 *NB East 86.32 NB 43.84 NBT 104.1 14.18 NB West 103.77 SB 42.06 NBR 104.1 13.55 SB East 82.20 EB 15.42 SBL 6.9 30.97 SB West 89.98 WB 13.76 SBT 167.1 8.43 EB North 21.11 SBR 167.1 10.99 EB South 21.75 EBL 11.5 12.90 WB North 25.76 EBT 91.9 4.51 WB South 22.05 EBR 91.9 3.46 WBL 7.3 20.67 WBT 50.8 4.58 WBR 2.2 9.51 *NB East denotes NB travel on the eastern most crosswalk

PAGE 117

117 Table A 40. Alternative, semi actuated, high veh, high bike, high ped. Signal System Type Semi actuated Vehicle Average Speed (mph) 11.56 Vehicle Volume Category High Weighted Veh Average Delay (s/veh) 8.99 Pedestrian Volume Category High Weighted Ped Average Delay (s/ped) 47.41 Bicyclist Volume Category High Weighted Bike Average Delay (s/bike) 26.66 Geometric Layout Alternate Cycle Length (Critical) (s) 192 Vehicle Approach Average Queue Length (ft) Average Delay per Vehicle (s) Pedestrian Approach Average Delay per Pedestrian (s) Bicyclist Approach Average Delay per Bicyclist (s) NBL 164.0 36.81 *NB East 89.23 NB 38.56 NBT 164.0 15.75 NB West 93.31 SB 41.97 NBR 164.0 16.84 SB East 92.60 EB 15.56 SBL 8.0 39.95 SB West 90.56 WB 15.58 SBT 181.0 9.36 EB North 25.26 SBR 181.0 11.13 EB South 25.24 EBL 13.9 19.32 WB North 25.68 EBT 100.4 4.87 WB South 24.81 EBR 100.4 9.88 WBL 8.8 26.80 WBT 54.5 5.01 WBR 2.9 16.33 *NB East denotes NB travel on the eastern most crosswalk

PAGE 118

118 Figure A 9 Alternative configuration on VISSIM for pre timed. Figure A 10. Alternative configuration on VISSIM for semi actuated

PAGE 119

119 LIST OF R EFERENCE S AASHTO (American Association of State and Highway Transportation Officials). (2012). A policy on geometric design of highways and streets Washington, D.C. AASHTO Executive Committee. (2012). Guide for the development of bicycle facilities, American Association of State Highwa y and Transportation Officials, Washington, D C And < http://peopleforbikes.org/blog/americas first protected intersection is open in davis and working like a charm/ > Department for Transport. < https://www.gov.uk/guidance/the highway code/rules for pedestrians 1 to 35 > Eco Counter (2017). CITIX IR System Overview < http://www.eco compteur.com/en/products/citix ir > Elefteriadou, L., Chase, T., Zheng, Y., and Kontou, R. (2015). Before and After Implementation Studies of Advanced Signal Control Technologies in Florida University of Florida Transportation Institute, Gainesville, Florida. Fong, G., Kopf, J., Clark, P., Collins, R., Cunard, R., Kobetsky, K., Lalani, N., Ranck, F., Seyfried, R., Slack, K., Sparks, J., Umbs, R., and Van Winkle, S. (2003). Signalized Intersection Safety In Europe. US Department of Transportation Office of International Programs Washington, D.C. Furth, P., Yu, M., Peng, F., and Littman, M. (2014). Mitigating the Right Turn Conflict Using Protected Yet Concurrent Phasing for Cycle Track and Pedestrian Crossings Proc., Transportation Research Board 93rd Annual Meeting Washington, D.C. 2197. Gettman, D., Folk, E., Curtis, E., Kacir, K., Ormand, D., Mayer, M., and Flanigan, E. (2013). Measures of effectiveness and validation guidance for adaptive signal control technologies. US Department of Transportation, Federal Highway Administration Washington, D.C. Gilpin, J., Falbo, N., Repsch, M., and Zimmerman, A. (2015). Evolution of the Protected Intersection. Alta Planning + Design Portland, Oregon. Hu, Y. (2014). The Impact of Pedestrian Activities in Adaptive Traffic Signal Control System Operations. University of Pittsburgh Pittsburgh, Pennsylvania.

PAGE 120

120 Iteris (2017). PedTrax < https://www.iteris.com/products/pedestrian and cyclist/pedtrax > McKenzie, B. (2014). Modes Less Traveled Bicycling and Walking to Work in the United States: 2008 2012. US Department of Commerce U.S. Census Bureau, New York. Mimbela, L. E. Y., Klein, L. A., and United States. Joint Program Office for Intelligent Transportation Systems. (2007). Summary of vehicle detection and surveillance technologies used in intelligent transportation systems Federal Highway Adm inistration, Intelligent Transportatio n Systems Joint Program Office, Washington, D C. NACTO (National Association of City Transportation Officials). (2011). Urban bikeway design guide New York. Rhythm Engineering. < https://rhythmtraffic.com/results/ > Sanburn J. (2015). How Smart Traffic Lights Could Transform Your Commute < http://time.com/3845445/commuting times adaptive traffic lights/ > Stanek, D., and Alexander, C. (2015). Simulation Analysis of Intersection Treatments for Cycle Tracks Proc., 2015 ITE Western District Annual Meeting Las Vega s Nevada. TRB (Transportation Research Board). (2010). Highway Capacity Manual National Research Council, Washington, D C. Xiao M., Zhang, L., Hou, Y., and Chuan, S. (2013). An adaptive pedestrian crossing signal control system for intersection. Proc., COTA International Conference of Transportation Professionals Procedia Social and Behavioral Sciences, Shenzhen, China. 1585 1592. Zhao, Y., and Tian, Z. (2012). An overview of the usage of adaptive signal control system in the United States of America. Applied Mechanics and Materials 178, 2591 2598.

PAGE 121

121 BIOGRAPHICAL SKETC H Tyler Valila was born and raised in Gardner, Massachusetts where he graduated from Gardner High School in 2012. He completed his undergraduate studies at the University of Massachusetts Lowell and graduated Cum Laude in the spring of 2016 with a Bachelor of Science in civil engineering. Upon graduating, Tyler began graduate studies at the University of Florida. While simultaneously working for the University of Florida Transportation Institute and dedicating time to the Institute of Transportation Engineers, he graduated with his Master o f Engineering degree in the fall of 2017.