Development of a model for determining workzone illumination requirements during nighttime highway construction

MISSING IMAGE

Material Information

Title:
Development of a model for determining workzone illumination requirements during nighttime highway construction
Physical Description:
xiii, 218 leaves : ill. ; 29 cm.
Language:
English
Creator:
Kumar, Ashish
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph D.)--University of Florida, 1994.
Bibliography:
Includes bibliographical references (leaves 213-217).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Ashish Kumar.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001981542
notis - AKF8473
oclc - 31920325
System ID:
AA00003215:00001

Full Text








DEVELOPMENT OF A MODEL FOR DETERMINING
WORKZONE ILLUMINATION REQUIREMENTS DURING
NIGHTTIME HIGHWAY CONSTRUCTION














By

ASHISH KUMAR


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF
THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA












ACKNOWLEDGEMENTS


would like to extend my deep gratitude to Dr. Ralph D. Ellis, my supervisory


committee chairman, for his guidance and dedicated support that made this study possible.

I am sincerely appreciative of Dr. Paul Y. Thompson, Dr. Charles Kibert, Dr. B.L. Capehart,


and Dr. J.


Wattleworth for serving on my supervisory committee.


I wish to thank Mr. Scott Amos, a departmental colleague and graduate student, for


his ideas and suggestions.


Most of all I would like to thank my wife, Akanksha, for her


unconditional support and enthusiasm during the long evenings and weekends of work


required


for this study.


Finally, much gratitude is owed to my parents for their love,


continuous encouragement and support during the entire course of my studies.















TABLE OF CONTENTS


ACKNOWLEDGEMENTS.

LIST OF TABLES . .


LIST OF FIGURES.

ABSTRACT .


. 11


S.
. 9 9 5 9 ,11


. .xUii
9.d


CHAPTERS


INTRODUCTION AND PROBLEM STATEMENT


Overview of Nighttime Highway Construction
Need for Adequate Lighting . .


Problem Statement.
Research Objectives.
Research Methodology


. 7
. 8


OVERVIEW OF STATE-OF-THE-ART


Introduction .


9 4 9 S 9 .1


9 9 S S 4 9 .1


Review of Literature. .
Evaluation of Visual Task


Luminance.


. . 13


Contrast


Reflectance.


Design Criteria. . .
Effect of Human Factors. . . .
Adaptation . . . .
Glare sensitivity . . ..
Physical & psychological considerations
Safety and accidents . .
Performance and productivity .
Lighting Standards and Guidelines. . .


. .21


S. .22
S9 2 .23


9 5 5 S S 2'1


1


,,


Parre








State DOT's


Current Practices


Specifications.


.39
.45


Nighttime work .
Lighting Sources and Equipment
Field Investigation .


. .45


. .48
. .49


Summary


MODEL COMPONENT


AND THEIR FORMULATION.


. . .56


Introduction.


Nighttime Work Activities


Preliminary Identification.


Factors


. .57


Summary of Survey Questionnaire.
Compilation & Analysis of Responses
Influencing Illumination Requirements.
Human and Cognitive Factors .
Environmental Factors .
Task Related Factors. . .
Lighting Factors . .
Identification of Significant Factors ..
Determination of Factor Levels .


. .69


. .75
..76


Non-highway Task Matrix.
Description. .
Sources of Information.


Development of Matrix and SAS Dataset
Summary


..78


MODEL DEVELOPMENT


. .81


Introduction .
Model Approach.
Regression Models


. .84


Simple linear regression


Least square method
Correlation. .


* 9* 4 9
* 9 4 t ft 9


S 4 9 .8


..87


Multiple regression and general linear model
Confidence intervals and prediction limits.
Criteria for Model Evaluation . . .
Residual analysis. . .
Lack-of-fit test in linear regression. .
Coefficient of determination . .


9. 8


..90
..92


S4 9 9 .9 5


56


. .76


.78


89








Fundamental Assumptions. .
SAS Procedures. .
Database Development . .
Data Analysis . . . .
Correlation Coefficients .
Trial Model Formation. .
Regression Analysis . .
Testing and Model Adequacy
Model Evaluation and Checking Assumptions
Summary . . . . .


* 9 9
* 9 9 9
* 9 9 9
* 9 9 9 9
* 9 9 9 9
* 9 9 .


S. . .98
S 99
. . 102
.104
. .104
. .* 106
* 9 9 109
.* 9 9 9 110
.* 9 9 9 122
* 9 9 128


DEVELOPMENT OF GUIDELINES.


9. . 130


Introduction . . . . ....
Development of Illumination Level Categories. .
IES Categories . . . . .
Nightwork Illuminance Level Categories. .
Recommended Illumination Level Categories for Nightwork
Developing Equipment Guidelines including Glare Criteria.
Consideration of Glare Criteria. . .
Summary *


MODEL APPLICATION CASE STUDIES


Introduction . . .
Case Study 1. . .
Description. . .
Tasks Identification
Level Determination
Comparison and Evaluation
Case Study 2. . .
Description.
Tasks Identification
Level Determination
Comparison and Evaluation
Case Study 3. . .
Description. . .
Tasks Identification
Level Determination .
Comparison and Evaluation
Summary . . . .


* 9 9 9 1
* 9 9 9 1
* 9 1.
Activities 1
* 9 9 9 1
* 4 9 9 1
* 9 9 9 9 1


. 153


.* 9 9 9 9 9 . 4 9 1
* 9 9 9 9 9 9 9 9 9 4 9 9 9 1





* 9 9 9 9 9 9 9 9 9 9 9 9 9 1










* 9 9 9 9 9 9 9 9 9 9 9 9 9 1


fly n nrn frlt AtI fltrl TY%%r 1'fl\Trr rrn r


lrl







Conclusions .


Suggestions for Future Research.
Glare Control .
Lighting Configuration


. 174
. 176


* . . 176
* . . 1 7 6


Additional Factors Affecting Lighting Requirement.


APPENDICE
A
B


. 177


S
QUESTIONNAIRE SURVEY AND FIELD OBSERVATION FORM 1


SAS SOURCE CODE.


SAS OUTPUT.


SUMMARY OF FACTOR COMPARISON FOR LEVELS.


REFERENCES.


. 213


BIOGRAPHICAL SKETCH


. 218







LIST OF TABLES


Table


Conditions for Discomfort Glare Acceptability.


..26


Relation Between Illumination and Night/Day Accident Ratio.


Illuminance Levels for Safety


Recommended Levels for Uniformity Ratios and Construction Activities


. .34


. .36


Minimum Illumination Intensities for Construction Industry.

Proposed 30 CFR Regulations Illumination Requirements for
Draglines, Shovels, and Wheel Excavators. . .

Summary of Provisions for Lighting Requirements and Guidelines


for Various States.


*4c0


Technical Information and Rating of Light Sources


Description of Highway Construction and Maintenance Tasks
Performed at Night . . . . .


Categorization of Typical Highway Construction Tasks and Operation


Summary of Questionnaire Survey of State Highway Agencies


..59


. .61

. .62


Summary of Questionnaire Survey ofFDOT District Offices


Number of States Performing Various Nighttime Highway Construction Tasks

Number of States Performing Various Nighttime Highway Maintenance Tasks


Performing Frequency of Various Construction Tasks on FDOT
Nighttime Projects . . . . .


. .65


Performing Frequency of Various Maintenance Tasks on FDOT
Nighttime Projects. . . . . .


. .66


3.9 List of Factors Significantly Affecting Nighttime Highway Task Visibility. .74


Page


.29







Table

3.11


Page


Factor Description and Illuminance Levels of Outdoor Industrial


Tasks and Spaces


Summary of Factor Subjective Levels and Corresponding Numeric
Value Assignment.


S103


Summary of Results of Coorelation Analysis.


Results of Correlation Analysis for Fixed Levels


. 104


of Importance and


Accuracy Factor.


Forward Selection Procedure for Dependent Variable LEVEL (steps 1 & 2)

Forward Selection Procedure for Dependent Variable LEVEL (steps 3 & 4)


111


Forward Selection Procedure for Dependent Variable LEVEL (step


Maximum R-square Improvement for Dependent Variable (steps 1 & 2).

Maximum R-square Improvement for Dependent Variable (steps 3 & 4).


Maximum R-square Improvement for Dependent Variable (step


Results of Different Regression Models


Predicted and Residual Values of LEVEL.


IES Illuminance Categories Recommended for Tasks and Spaces


. 113


.114

1. 15

. 116


.123


.125


. 132


Recommended Illuminance Ranges and Categories for Nighttime Highway


Construction and Maintenance Tasks


Suggested Illumination Categories and Levels for Typical Highway
Construction and Maintenance Tasks . . .


. 139


Recommended Illuminated Area in the Direction of Travel for


Various Construction Equipment


. 142


Utilization of Glare Avoidance Screens and Barriers by Various States


. 147


134







Task Identification for Barrier Wall Replacement Operation .

Task Identification for Repaving Intersection Operation.


. . 155


. 162


Task Identification for Bridge Deck Construction Operation.


.168







LIST OF FIGURES


Summary of Factors Affecting Night Operations.


Schematic Flow Chart of Research Work Plan. . .

Retroreflectivity Coefficients as a Function of Distance for Various


Road Surfaces.


Comparison of Various Lighting Design Criteria. .

Relation Between Illuminance (E) and Luminance (L) for the
Surface of Diffused Reflectance. . . .

Effect of Glare Source on BCD Luminance at Different Levels


Background Luminance.


Performance as a Function of Illuminance for Different Task Sizes


and Varying Contrast


Relation Between Performance Speed and Visibility Level.


. .32


Methods of Measurement of Luminous Intensity.


. .38


Summary of Factors Influencing Task Illumination Requirements


..68


Flow Chart of Model Development Procedure.


Various Regression Methods


Graphical Representation of Least Sqaure Method.


. .88


Graphical Representation of Prediction Limits.

Some Possible Patterns of Residual Analysis.


Plot between Level of Illumination and Factors


. 107


Residual Plot of Illumination vs Importance.


117


- S t a ---a-


igr


. 16


3


Page


82


.r







Eigvre


Residual Plot of Illumination


vs Importance, Speed & Size


. 119


Residual Plot of Illumination


Residual Plot of Illumination


vs Imp, Speed, Size & Distance


vs Importance


.120


All Five Factors .


Plot of Residual and Predicted Values .


.126


Distribution of Residuals


.127


Results of AGI Simulation for Lighting Configuration of


Slow


Moving Equipment


Results of AGI Simulation for Lighting Configuration of Fast
Moving Equipment . . . . .


.144


.145


Effect of Beam Angle and Eccentricity in Reducing Glare.

BCD Luminance for Glare Sources at Various Angular Distance


. 148


from the Line of Sight.


Plan Layout of Workzone (Case Study 1: Bar

Perspective View of Workzone (Case Study 1


Plan Layout of Workzone (Case Study


Perspective View of Workzone (Case Stu'


. 149


rier Wall)


Barrier Wall)


Repaving Intersection)

dv 2: Reoavine Intersec


S


Au


.157


. . 160


:tion)


. 161


Plan Layout of Workzone (Case Study 3: Bridge Deck) .


Perspective View of Workzone (Case Study 3: Bridge Deck)


. 167


Page












Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DEVELOPMENT OF A MODEL FOR DETERMINING
WORKZONE ILLUMINATION REQUIREMENTS DURING
NIGHTTIME HIGHWAY CONSTRUCTION

By

Ashish Kumar


April 1994


Chairman:


Dr. Ralph D. Ellis


Major Department:


Civil Engineering


With


the increasingly popularity of nightwork


on many


highway


construction


projects, a number of complex problems have been introduced in highway rehabilitation.

Of many issues involved, workzone lighting is found to be the single most important factor


affecting quality, safety, cost and productivity on these night projects.


At present, lighting


requirements in the workzones are minimally defined in the specifications of few state


highway agencies and

guidelines to go by. Alt


often times are not formally designed as there is no standard or


hough a number of research studies have been performed in roadway


lighting and industrial lighting area, no formal study determining illumination requirements


in the workzones has been conducted.


In this research effort an attempt has been made to


identify the typical highway construction tasks performed at night and the factors affecting







determining average maintained light intensity levels for the tasks.

In an effort to provide an overview of nighttime highway construction and provisions


for lighting,


an extensive literature review was conducted.


As a result of nationwide


questionnaire


survey,


various


highway


operations commonly performed


at night were


identified. Significant factors influencing illumination requirements for workzone lighting were


also identified.


These factors included


- a) speed, b) accuracy or importance, c) reflectance,


d) seeing distance, and e) size of object.


These factors were assigned certain subjective levels


for comparing with visually similar outdoor industrial tasks.


Various illuminance level


categories were determined based on IES and OSHA recommendations, state provisions,


opinions of experts and experiences during field reviews.


The categories were intended to


interpreted


as recommended


safety


requirements


not regulatory


minimum


requirements. A model approach was adopted to determine illumination levels for any given

construction task. Correlation analysis was performed to examine the association of various


factors with illumination requirement.


procedure

studies.


s.


Trial models were suggested and analyzed using SAS


The most appropriate model was selected and validated using three real life case


For all the case studies which represented most commonly performed highway


construction tasks at night, the results conformed to the field recorded values.


opinions of the crew at the job-site and experiences while recording the observations, it was

concluded that the results from the model were in agreement with the findings.


From the












CHAPTER


INTRODUCTION AND PROBLEM STATEMENT


Overview of Nighttime Highway Construction


Recently many states have changed the direction of their operations from new


highways and roads to maintaining the existing ones.


The emphasis has shifted from


building new facilities to maintaining and improving those in existence.


many problems.


This shift creates


One such problem arises from daytime lane closures which results in heavy


congestion on roads already loaded to capacity.


This problem is not limited to roads in


urban areas, but also includes some rural highways that are often as crowded during certain


times of the year as urban areas.


According to a state highway agency official, it has


become hard to separate the morning rush hours from evening rush hours and congestion


lasts for 12-13 hours a day.


This creates the situation where the natural, ordinary solution


of lane closure becomes unrealistic or impossible during peak times. The daytime lane

closures are also hazardous, costly and inconvenient for the traveling public. As a result,


more construction and rehabilitation work is being performed during hours when traffic flow


is minimal.


For this reason many highway agencies have started working at night.


addition to several obvious advantages of nighttime work such as cooler temperatures for

equipment and material, less traffic problems, and delays, there are certain complex issues










A recent research study


conducted at the University of Florida performed an


extensive literature survey and discussed various issues related with nighttime highway


construction (1).


The literature review from the study on highway construction also


confirms this trend of increased nightwork and addressed the problems associated with it.

However, the number of references dealing directly with the night shift operations, as a


whole, are limited.


Only a few studies provide a comprehensive approach and valuable


information towards night shift construction.


In the study based on the literature survey,


analysis of several case studies, and meetings with experts in Florida and other states, a

number of factors were identified that affect the decision of shift times for highway


construction.


night.


These factors play important roles in determining whether or not to work at


Based on their characteristics, they are divided into five categories and presented in


Figure 1.1.


A listing of the categories with respective factors follows:


1) Construction related factors

a) cost, b) productivity, c) quality, d) noise

2) Traffic related factors

a) congestion, b) safety, c) traffic control

3) Human factors

a) sleep, b) circadian rhythms, c) social/domestic issues

4) Miscellaneous factors

a) public relation, b) information, c) supervision,
























C.
Lie 0


c0 BC cn

(A 0,




I I aoao u,Z Cr~oZ




r* -
m CT C
P~ o 4-

SL o LL
ac _) 0 r:




__o I

a c(A




O C
o 4-C4-
U, CO -0






St
We =
Otj V
ti P







Some of these factors have qualitative and some quantitative attributes.


For instance,


cost and productivity are quantitative factors; quality, noise, safety, and congestion are both

qualitative and quantitative; and human factors and other factors are qualitative. As a result

of analysis of construction related factors, it is found that cost, quality, and productivity had


some project-by-project variations due to various reasons.


Cost, productivity and quality are


the basic project attributes and are affected by the way work is performed.


There have been


various opinions with regard to variations in these factors as a result of a change in workshift.

According to some opinions, construction costs and total project costs are normally expected


to be higher in the case of night construction as opposed to day construction.


are attributed to overtime, shift differential, lighting, higher bids etc.


public costs may follow a reverse trend.

on quality and productivity. The numb


The reasons


However, user and


Similarly nighttime work may have negative effects


er of noise complaints are higher for nighttime work


if projects are in urban areas and take a longer time to finish.


Although human factors are


usually given some consideration, they do not constitute a major decision-making criterion.

Traffic related factors are considered to be the most important when deciding about night


work.


Congestion is often found to be the one single factor resulting in the decision to use


a night shift on many projects.


Safety is one of the major concerns during night work because


accidents tend to be more severe, even though the rate is relatively low.


To enhance safety


appropriate traffic control is emphasized and slower speeds, lane closures, detours, use of

flashing arrow boards, use of appropriate warning devices, layouts, sufficient lighting and use

of police patrol cars are found to be some of the more effective measures for better traffic








but the public has a positive attitude towards the delays and noise.


Supervision and


communication are difficult for nighttime work because most offices are open during the


daytime.


Similarly, supply and repair also create some problems since parts and materials are


not easily available at night.


Although most of these factors have some effect on night work,


work zone lighting has a considerable influence.


It not only affects quality and productivity,


but also influences traffic control, safety and human factors.


Need for Adequate Lighting


A study conducted at the University of Florida has shown lighting as one of the most


important factors for nighttime construction (1).


It has been found that safety in the


workzone,


traffic control,


quality of work and worker's morale are directly related to


workzone lighting.


Limited, or restricted visibility is an obvious drawback of nighttime


construction.


However,


with adequate levels of light,


construction operations can be


performed as well at night as they can be during the day.


Likewise, the key factors that


influence accident rates are the physical conditions of drivers and the light conditions of the


environment.


The physical conditions of the driver may include conditions such as drowsiness


and sensitivity to light glare.


Worker injury rate also increases due to the inherent vision


impairment associated with nighttime lighting conditions (2).

area is also important from the point of view of quality. Sta


Sufficient lighting of the work


ndard highway lighting, or light


from nearby businesses or residences, is generally inadequate to properly light the areas where


work is performed.


Inadequate lighting results in problems with proper inspection.


Work











Problem Statement


In the specifications of many state highway agencies present lighting requirements are

minimally defined (e.g. minimum intensity level of 5 ft-candles, sufficient light to permit good


workmanship and proper inspection).

but also not standardized. All the de


discretion of the site engineer and the contractor.


Most of these specifications are not only inadequate


'cisions pertaining to workzone lighting are left at the


Moreover due to the lack of criteria related


to average illumination levels and uniformity of illumination, lighting systems can not be

designed satisfactorily.

Various question arise related to this issue which include the following:


What should be the definition, size and dimension of workzone area during


nighttime work ?

2) How far should the view of the construction equipment operator be extended to

satisfactorily perform the task?


What


is the


minimum


illumination


requirement


adequate


visibility


comfort?

4) What are the factors that determine quantity and quality of light for any particular

task?

5) How can the average illuminance for typical construction and maintenance tasks

be determined?




7


7) What is the minimum illuminance level required for work inspection and quality

control?

8) What should be the extent of lighted area around the equipment for the workers

to work comfortably ?

9) Is there any need of portable or fixed lights in the workzone?

10) How can the discomfort glare to the motorists and the workers be reduced and

avoided?

These questions warrant the need and importance of a study which can provide


satisfactory approach to solving such problems. Although lighting is the single most

important factor in nighttime construction, it is least studied. Very little research has been

done concerning the proper lighting of construction sites. Although a number of studies


concerning various lighting aspects such as industry lighting, roadway lighting, and building


lighting are conducted.


They are considerably different from construction site lighting and


can not be directly applied.


Research Objectives


In the light of previous discussion it becomes imperative to conduct a research effort

which not only can suggest the design of illuminance levels based on current site conditions,


but also facilitate developing set of standard guidelines and specifications.


The main objective


of this research effort is to develop a model approach to determine the illuminance levels for

nighttime highway construction tasks and to develop and recommend guidelines for these




8


1) To review the fundamental lighting concepts from literature, state-of-the-art of

lighting, and current practices in nighttime construction. The intent is to determine the typical


highway construction and maintenance activity list commonly performed at night.


This also


includes determining the factors that affect illuminance requirements for a particular task and

assigning levels to these factors for numerical comparison purposes.

2) To develop a regression model which can best fit the data of existing standards and

can be used as a prediction model for highway operations.


3) To develop the guidelines by


comparing highway and non-highway tasks which


consists of illumination categories, recommended levels for typical tasks and equipment.

4) To validate the model by analyzing several real life case-studies and comparing the


results with the ones observed in the field.


Suggestions will also be provided for model


modification, improvements and future research.


The detailed research work plan has been provided in the subsequent section.


Figure


1.2 shows the schematic flow chart, which explains exactly how the above stated objectives

are accomplished.


Research Methodology


accomplish the objectives of this study several


tasks were completed.


comprehensive review of published information in the study area was conducted.


Literature


pertaining to visual task evaluation, and role of illuminance, contrast and reflectance in

visibility of objects were collected from all relevant U.S. and international electronic data





























RESEARCH METHODOLOGY I


Figure 1.2


Schematic Flow Chart of Research Work Plan




10


Principal equipment manufacturers and lighting manufacturers were contacted to


obtain a summary of information concerning available lighting equipment.


was prepared to be sent to all state highway agencies.


A questionnaire


The purpose of this survey was to


obtain information concerning the range of construction and maintenance activities performed

at night, and illumination criteria used by different state highway agencies.

A number of nighttime highway construction sites were visited to collect data about


current


practice.


These


visits


served


to document


information


about


configuration of lighting equipment in use, light intensity levels currently being obtained, and


workers' perspectives.


This field survey involved video taping of various work activities


being performed at night and


measurement and recording of light intensity levels.


Field


reviews were targeted at the types of construction and maintenance activities most commonly

performed at night.

Identification of various factors which have significant influence on determining


adequate illuminance levels was done.

literature and experience of experts. Varin


This identification was based on the review of


ous subjective levels were assigned to these factors


to develop nighttime lighting requirements for highway workzones by transferring relevant


standards


from


similar


non-highway


activities.


was


accomplished


primarily


determining the


"seeing tasks"


worker needs for the highway activities and


identifying non-highway activities with similar "seeing tasks" and worker needs.

Using SAS software, correlation analysis was performed to evaluate the level of


significance of various factors.


The factors having high correlation were selected for model






as the regression equation for future predictions.


This model was then tested by examining


three undergoing nighttime projects in the


state of Florida.


The model was applied to


determine the levels for typical tasks performed in each operation and compared with the


actual recording of levels and workers' opinions.


These case-studies validated the model


developed for prediction of illuminance level for any nighttime construction activity.


set of


guidelines


were


recommended


typical


construction


maintenance tasks identified earlier.


For the guidelines, several illumination categories were


developed and illumination ranges were recommended.

approach suggested for lighting in other industries. Rec


These categories followed the


:ommendations for improving and


modifying the model have also been included as suggestions for future research.












CHAPTER


OVERVIEW OF STATE-OF-THE-ART


Introduction


In an attempt to provide an overview of state-of-the-art, an extensive literature survey

has been conducted to obtain the most up-to-date and relevant data pertaining to illumination


construction workzones.


computerized search of


electronic databases including


Southern Technology Application Center (STAC) and Transportation Research Information


Services (TRIS) resulted in more than 100 published papers and articles.


Additional reference


materials were obtained through the Illumination Engineering Society of North America


(IESNA) and the International Commission on Illumination (CIE).


Although no articles were


found that directly addressed the topic of construction workzone illumination, many were

found in related areas such as roadway lighting, industrial lighting, surface mine illumination,


sports lighting, and the fundamentals of visibility.


The literature fell into four basic areas:


visual task, human factors, current practices and standards.


In addition, literature review


information pertaining to standards and guidelines was also collected through a questionnaire


survey. Similarly, field investigation and survey contributed to the additional data on current

practices. It should be noted that although the variables are very often discussed individually,


a large number of inter-relationships exist.








Review of Literature


Based on the literature survey, it was determined that visual perception depends on


a number of factors. The important factors include

1. Luminance in visual field

2. Contrast

3. Reflectance of the object and background

4. Physical properties of objects

5. Duration of exposure

6. Physiology of the observer's visual, neural and mental systems

7. Age

8. Psychological and cognitive factors

Of the above factors, the ones related to human issues such as adaptation, physio-

logical and psychological considerations, and effect of physical properties of objects in


performance and safety are discussed in the section addressing effects of human factors.


remaining factors are included in the section addressing evaluation of visual task.


Evaluation of Visual Task


Visual displays can be specified in terms of quantities of various parameters, which

include: 1) luminance, 2) contrast, and 3) reflectance.

Luminance


A P ^; ;i nnrnot"


Tnrrrd tlrm~nnn~a Zo nn ;mnn~n*t nnnmlllnn SnFtnr ;n ttf ollnl nFll;)rr








increased.


Results of research by Smith and Rea also confirmed that a certain minimum level


of illumination is required for optimal task performance (4,


However, exceeding this


minimum by a significant amount will not cause performance to deteriorate, as long as glare

is minimized or avoided.


Finch


also emphasized


luminance is one physical


parameter that


must


prescribed as the basic quantity in the visual field (7). Luminance may cause a reaction within


the observer's visual system that results in a subjective response termed "brightness."


addition to quantity considerations, the quality consideration associated with luminance is


related to uniformity of luminance.


is very important.


With regard to seeing detail in the work area, uniformity


Uniformity, depending on the requirement, can be measured in three ways:


1) minimum luminance in the visual field compared with the adaptation luminance, 2) the total

luminance variation in the visual field, and 3) ratio of maximum and minimum luminance.

Although luminance is the parameter that directly affects visual perception, in addition


to luminance values, a set of illumination values should also be prescribed.


The primary


reason for this is that the collective experience of roadway and other lighting designers in the

country relates mainly to horizontal illumination values which makes the practitioners


comfortable with a specification that requires known quantities (7).


The relationship between


luminance and illumination and their respective advantages are discussed in detail later in the

section addressing various design criteria.

Contrast


Contrast is usually defined as the ratio of the luminance difference of target and








does not change by varying intensity of the source.


However, since reflectance depends on


the positions of observer and source, changing the position of the source may change contrast

(8).


As the contrast of the object is raised, the probability of seeing increases until it can


be detected 100% of the time. The contrast at which the object can be detected 50% of the

time is called the threshold contrast. A fundamental relationship exists between object

detection and the luminances of the background. According to IES, as the background


luminance increases the contrast threshold decreases with a decreasing rate (8).

Ginsburg found that contrast sensitivity, and not visual acuity, helps predict an

individual's ability to see an oncoming target or stationary object at the first possible moment


Gallagher and Meguire in their study of contrast in night


correlation with driver performance (10).


driving found a strong


Contrast was found to be the most important


element in determining visibility within the range of variables studied.

Reflectance


Reflectance of a surface is defined as the ratio of its luminance to the incident


illumination.


For highways, two types of reflectances are considered: 1) retroreflectance and


2) forward reflectance.


Retroreflectance is defined as the percentage of lamp illumination


returned by the highway surface to an observer on the same side of the lamp.


Forward


reflectance is the percentage reflected by the surface to an observer on the opposite side of

the lamp.


Figure


2.1 shows the retroreflectance data for various road surfaces (11).


In general,





16





oC
o 4






1V
\ \ <
/ 0
/ C-
N 0





0) U O
4- U 0 0)




o C


a a
o 0t



o 'U-O


v, t
Na


eI-




0< O
4- O



S O o
Q) 4)






>a
V0 C






00




1d O
Ow W LLe

O /' *
2\' I '- -*
S o









O
0 0 .\ **











0 Q
HC 0
-* \ o









v O .
HV \
o \)
w- 'N 0








(0 0
-c '* \(NjLU



6\ 6 0
\ *-0i



== \ } I

>- ;n 0) \I :
t: :: \




17


However, it does not vary significantly with the lateral position of the measured point beyond


a distance of 100 ft.


Forward reflectance values are usually 10 to 100 times greater than


retroreflectance values because of a large specular component.


Maximum values lie along the


source-observer axis and peak at a point that lies between the lamp and the midpoint of the

axis.


Design criteria.

The objective of lighting design is to provide the proper quantity and quality of light


based on several economic and practical considerations. However, quantity and quality of

light required for a particular task is governed by the design criteria. As pointed out in IES,


in the past there was an overwhelming emphasis on delivering sufficient quantity of light on


the task or work surface (8).


Quality of light has been considered mainly in controlling direct


and reflected glare, and uniformity.


However,


with advancements in lighting research, a


systematic change in the objectives of light design and requirements was observed.


In 1982,


the Roadway Lighting Committee ofIESNA approved a revision to the Standard Practice for


Roadway Lighting (12).


This revision incorporated design criteria based on pavement


luminance, luminance uniformity and veiling glare as the preferred and equally acceptable


method for roadway lighting.


The old method was based solely on horizontal illuminance.


The three major design criteria are: 1) illuminance criteria, 2) luminance criteria, and

3) visibility criteria.


Figure


depicts the differences among the three criteria.


Each of these criteria


requires determination of different physical variables and differs in terms of expected accuracy





18









~~~~~ -~a, 1
H- >4 -4--~ 0
-C >0 0- -du~~r
rflLd *00>
,rrW La- Q
con: p)X ~,
-Y .-Wcn CU
(000Qc 00000. a




Cd
'Zn


C

a- '4-



cc c Cr CU4
-r cr .- Va
Do 0) 0d 0. O w-r c
>CT -- aJEC) cncr
~W re CCU
Q04
a~ca err a, cc S




a I-= m (da


w 4- E



4) 4-
nO. -c a
'4
zwC 4-C 00~

Ec .r -.L
OC~~Z~ CO Q Q-c
-- U Uf U




19


source distribution information can result in an expected accuracy is +5%. The accuracy is

prescribed by the importance of the result weighed against the cost and complexity of the

calculation (8).

Illuminance criteria for a surface requires a certain quantity of light reaching that


surface directly from the source and by reflections from other surfaces.


Illuminance (E) is


defined as incident luminous flux (I) per unit area and is given by Lambert's cosine law as


= I cos O/d2


where 0 is the angle of incidence and d is the distance between the source and the point

(8). Although other criteria determine better visibility compared to illuminance criteria,


many regulations recommend illuminance levels for specifications.


Illuminance, despite its


limitations of determining accurate visibility, is more useful for design purposes as well as


quality


control


illumination


measurement


complex


computations (13).

Luminance criteria provides information regarding the appearance of objects and


surfaces,


their


contrast


luminance


ratios.


Determination


of luminance


requires


computation of illuminance on a surface and its reflectance characteristics.

directly proportional to the product of illuminance and surface reflectivity.


Luminance is

Weis has given


relationship


between


luminance


illuminance


for various


surfaces


diffused


reflectance as shown in Figure 2.3


(13).


The figure also shows the typical ranges of


reflectivity for various highway construction related objects.

Visibility criteria, which is a relatively newer concept, is an extension of luminance










(lux)


0.3


0.4


0.5


0.6


0.7


0.8


0.9


(Cd/sq.m.)




21


which reduces the perceived contrast of a complex target to threshold and thereby establishes


its visibility.


CIE published report No. 19/2 describes a mathematical model for influence of


lighting parameters on visual performance (15).


In the model, a term visibility index is defined


VI = (C.RCS. DGF.TAF)/0.0923


where


= contrast


RCS

DGF

TAF


= Relative Contrast Sensitivity

= Disability Glare Factor

= Transient Adaptation Factor


Physical variables required to determine these factors are:


1) luminance of task, 2)


luminance of the background around the task, 3) task size (in angular minutes),


observer, and 5) veiling luminance.

determined by these variables. Since


4) age of


All the above described factors, except TAF, can be


e research pertaining to quantifying TAF is limited and


available evaluation data is also not enough, TAF is usually disregarded during calculations.

Sometimes, however, its value is assumed between 0.97 and 1.0.


Effect of Human Factors


Human vision and perception adds another dimension to the complexity of illumi-


nation problem.


Understanding various human factors including physical, physiological and


psychological factors is crucial to evaluating its effect on productivity and performance under




22


physical and psychological considerations, and effects of human vision on perception and

safety are discussed in detail in following sections.

Among several visual functions and physical considerations, Henderson and Burg

indicated that adaptation and glare sensitivity are presumed to be of direct concern in night


work (16).


The primary reason is attributed to their influence on other visual functions


necessary in night driving and night work such as static visual acuity.


These factors in


addition to varying with the individuals, also depend on environmental considerations.

Adaptation


Adaptation is defined as a process of changing sensitivity of visual system to light in


order to detect the faintest signal.


Sensitivity increases under reduced illumination (dark


adaptation)


decreases


under


increased


illumination


(light


adaptation).


Schmidt


distinguished


three levels of adaptation corresponding to photopic (daytime),


mesopic


(twilight), and scotopic (night) vision (17).


The mesopic range, in which the illumination level


varies from


1.0 to


cd/sqm, is of primary interest in nighttime work and driving.


Transient adaptation is a phenomenon associated with reduced visibility after viewing a higher


or lower luminance than that of the task.


Recovery from transient adaptation usually takes


a fraction of a second; however, for slow recovery (i.e., more than one second) especially in


the case of a brightly lighted exterior, it can be a problem (8). During the process of transient

adaptation recovery any recognition task needs higher contrast. The greater the change in


luminance level is, the greater is the additional contrast necessary for recognition.

Recovery from transient adaptation usually results in momentary losses in visibility.







found at higher levels (18).


However, at low luminance levels, sudden decreases caused


smaller losses than those observed at higher levels.

Glare sensitivity


Glare is usually classified in two categories: 1) disability glare and 2) discomfort glare.

Disability glare, which is also known as veiling luminance, is caused by scattering of incident


light by the ocular media.


Because of this scattering every luminous point in space acts as a


source of stray light for nearby points resulting in mixing of images on retina and reducing


contrast.


Discomfort glare, which is also called as direct glare, is a sensation of annoyance


caused by viewing a high brightness source directly.


Discomfort glare can be reduced by de-


creasing the luminance or area of light source or by increasing the background luminance

around the source.

A study on glare recovery reported a progressive deterioration in glare recovery ability


with advancing age and the rate is greatly accelerated after age 40 (19).


In general, recovery


from the effects of glare depends on the retinal area involved, its previous adaptation level,

intensity of the glare, exposure time, color of the glare source, accompanying changes in pupil


visual health of the individual (20).


In addition to its direct physical effects,


discomfort glare encountered in night driving and night work may be an important factor in

inducing fatigue and drowsiness, even if the glare is not strong enough to directly reduce

visual efficiency.


The method of evaluating disability glare was suggested by Holladay (21).


The effect


of glare is quantified by an equivalent uniform luminance that describes the effect of the stray







entire visual field.


To evaluate its effect, this value can be compared with the average


luminance or the adaptation luminance.


In order to specify limits of source luminances to


avoid disability glare, various studies have been undertaken.


One such recommendation from


CIE for mining applications accepts a value of 3000 cd/sqm in miner's visual field as

permissible (13).

In order to quantify and measure discomfort luminance, Luckiesh and Guth defined


a threshold luminance just necessary to cause discomfort (22).

borderline between comfort and discomfort (BCD). Invest


This threshold is called the


igations revealed that BCD


luminance increases linearly with the increase in background luminance for interior spaces and


night driving.


Figure 2.4 shows the effect of glare source size on BCD luminance, which


decreases with the increase in size of glare source at different of background luminance(8).

As a subjective appraisal of discomfort glare in roadway lighting, the CIE has sug-


gested a relative index known as Glare Control Mark (GM)(23,24).


scale indicating the degree of discomfort experienced.


GM provides an ordinal


The value of GM is related to different


glare sensation as follows:


Glare Sensation


Unbearable


GM-


Disturbing


GM-2


Just admissible


GM-5


Satisfactory restriction

Unnoticeable


GM-7

GM-9








10000
















1000
















100


BCD LUMINANCE (Cd/sqm)


DISCOMFORT ZONE






L 343 Cd/sqm



L\
V\

L 34.3 Cd/sqm


L 3.43 Cd/sqm


COMFORT ZONE


2.000E-04


0.002


0.02


0.2


SIZE OF GLARE SOURCE STERADIANSS)


Figure 2.4


Effect of Glare Source on BCD Luminance at Different Levels of


Background Luminance


Source: Ref 8





26


ciated value of the GM depends on the photometric and geometric characteristics.

A visual comfort probability (VCP) system has been developed to evaluate accept-


ability of lighting systems and its environment as being comfortable (25).


VCP indicates the


ratio of observers accepting the system as comfortable to the total observers when observers


are subjected to direct glare from the luminaries.


Table


gives conditions based on


consensus for which discomfort glare is not a problem.


Table


Conditions for Discomfort Glare Acceptability


Angle above Nadir

(degrees)


I I


Maximum

Luminance


(cd/m2


7710

5500

3860

2570

1695


Visual Comfort

Probability

(VCP)


70 or higher


Uniformity Ratio

(Maximum/Average

Luminance)


5:1 or less


Source: Ref 25


In a study of workers' sensitivity to glare Guth found that mining workers are exposed


to high degrees of glare, especially from the lighting equipment on machines (26).


As a result


thV art IaCC Cndtoiivu tn dlare anti clort* hahsr 1TITl hlmintnanec




27


Physical and Psychological Considerations

Determining adequate lighting not only requires physical considerations including task


visibility but is also affected by various psychological issues.


Several studies were conducted


to examine the "comfortable" level of illuminance and observer's preferences.


Bodmann


found that judgement of optimum illuminance increased with age and task contrast (27).


general trend of increased satisfaction with higher illuminance and decrease in satisfaction at

very high levels ofilluminance was also observed and confirmed by Boyce (28).

Some researchers have also studied the psychological effects of uniformity of light.


In general, one


sees


better with more uniform light distribution in the visual field.


Tuow


found


in an experiment that the subjects preferred uniformity ratio of 3:1


for higher


illuminances (500 Ix) whereas for lower illuminances a ratio of 2:1 was preferred (29).

Research has also been conducted on human factors and behavioral studies to examine


behavioral changes with lighting levels.


Taylor concluded that lighting affects spatial


orientation of people and when navigating around a barrier, people tend to follow the


direction of higher illuminance (30).


Another study investigated subjects' preferences as


related to brightness and found that subjects preferred to work in the areas facing or near

bright areas (31).

Lighting was also found to have a psychological affect on people concerning their


activity level and attention not directly related to seeing and vision.


Activity level seemed to


be increased with higher illumination and similarly reduced with low illuminance levels.


Light


was found to have an effect on the performance of a reaction time task (32).







Safety and Accidents


The need to conduct highway maintenance and reconstruction creates legitimate safety


concerns, for both the workers and the drivers. The issue of safety during nighttime

construction has been addressed in several research reports. One such study listed some


limitations of nighttime construction influencing safety, which included (1):

1) Light conditions of the environment and visibility;

2) Glare and disorienting effects of sudden bright lights;

3) Reduced level of alertness on the part of the work force due to working alternate

shifts and fatigue;

4) Physical condition of drivers such as drowsiness, sleepiness, fatigue and overdrive,

etc.;

5) Higher speeds on the highways due to decreased traffic volumes; and,

6) Greater incidence of drivers under the influence of alcohol and other substances.

According to a California Department of Transportation (DOT) report nighttime


construction


accounted


percent


of the


injuries/fatalities


sustained


highway


workzones (33). Most worker fatalities (and major injuries) are related to errant drivers

causing these accidents. One study of psychological behaviors pointed out that most


confident drivers tend to "overdrive" the illumination afforded by the headlights of the vehicle


(34).


With darkness, various recognition functions such as activity, sensitivity to contrast and


ability to perceive objects degrade rapidly.

Standing Committee on Highway Traffic Safety of the American Association of State








and found nighttime accidents were more severe than daytime accidents (35).


The study


concluded:

workzone fatal accidents concentrate in rural areas and the vast majority of
all workzone accidents and injuries are concentrated in urban areas. .
Accidents occurring in workzones are generally more severe producing more


injuries and fatalities than the national average for all accidents
two-thirds of all accidents in the workzones occur in daylight.


occurring in darkness are far more severe.


S. more than
The accidents


Nighttime accidents account for


more than half the fatal accidents and more than their proportionate share of


injury accidents.


darkness.


About half of all workzone fixed object accidents occur in


p.16-17)


In an effort to reduce night accidents on a street, a four-year study was conducted and


illumination levels were maintained at a design level of 1


Ix (36).


Accidents at night were


found to be reduced due to improved illumination of the street.


A study undertaken by the


Illuminating Engineering Research Institute gathered data from six metropolitan areas to

study optimum illumination levels and uniformities from the standpoint of accident reduction


(37).


Although the study did not conclude that higher illumination levels necessarily reduce


accidents, it also did not reject the possibility.


The findings are summarized in Table


general, streets with little or no illumination had substantially higher night-day accident ratios.


Table


Relation Between Illumination and Night/Day Accident Ratio


Illumination Level Night-Day Accident Ratio
(in fc All Accidents Injury fatal
0.3-0.6 1.07 1.40
0.8- 1.1 1.69 1.93





30


Performance and Productivity

The IESNA has recommended illuminance levels for various tasks for nearly eight


decades.


The objective of which has been to maximize productivity in terms of greater speed


and accuracy while limiting the cost of lighting equipment.


It is recognized that, in general,


visual systems perform better at higher illuminance levels (8).

Several studies have been conducted imitating realistic tasks to determine how illumi-


nation affects performance.


One of the earlier studies indicated that as background luminance


increased, performance measured in terms of speed and accuracy, increased (38).


increase had a trend of diminishing returns which was more visible for high-contrast, large


targets than for low-contrast, small targets.


Figure


5 shows the trend for varying contrast


and target sizes.

Rea and Smith performed several studies on more realistic tasks under different levels


of illumination (39).


In a needle probe experiment, Smith confirmed Weston's findings that


performance improved at a decrease in rate, as light level increased to about 940 fc (10,100

Ix). He concluded that

Performance was better with a white background than with a black one under


the same illumination.


However, for equal luminances or visibility levels,


performance for the black background was superior.


Figure


.6 shows the results of the study.


Other studies conducted by Rea and Smith


included experiments involving:


1) proof reading of paragraphs for misspelled words, 2)


reading for information taking a reading test, and 3) check value verification (5, 6,


It was


.......ll A~ C~1.. C*A LACAIIYA ...JI* :it. at a n~t a11 a.:IC al .. nenAAJ :, al I.. -H :u.. I ..l : ... 1 .: A.












0.97
0.5
| .056
0,4 ^ ^


0.4 k

00.3


S_0.28

C=,-


5 50 500


Illuminance (Ix)


5 50 500


Illuminance (Ix)


Target Size


= 4.5 min


Target Size


=3.0 mm


5000


Illuminance (Ixi


Target size


Figure 2.5


- 1.5 min


Performance as a Function of Illuminance for Different Task Sizes


C =


0~28
1 -' '









Performance Speed (inverse seconds)


0.08


0.07


0.06


0.05


0.04



0.03


0.02


0.01



0


5 10 15


Visibility Level (VL)


- ... I- O.-..a ....s i r:4k:1:+ r T i xr^ l




33


Lighting Standards and Guidelines


A combination of codes, regulations and standards affect lighting design and the


choice of lighting equipment.


These include state, national and international codes, local


ordinances, and professional and manufacturers standards.


The standards and guidelines are


intended to provide adequate lighting for various visual tasks, which sometimes are restricted

by local ordinances intended to reduce light spillage, restrict illuminance levels, and reduce


glare


However,


most current lighting standards are largely based on the illuminance


recommendations of the Illuminating Engineering Society (IES).


IES Standards.


Selection of appropriate illuminances was suggested by the IESNA following an


evaluation of factors affecting visual performance (8).


According to the recommendation,


illuminances for interiors are provided in the form of general categories and a specific


illuminance level from a category is selected based on the weight of certain factors.


ranges of the categories conform to the ones recommended by the CIE which were derived


on a consensus basis (41).


The factors playing important roles in the selection of a particular


level for interiors include: 1) age of the observer,


2) size of the target,


3) contrast of the


4) need for speed and 5) accuracy.


For exterior, however, fixed values ofilluminance


levels to be maintained on the task are recommended.

The minimum required illuminance levels from safety considerations are shown in


Table 2.3.


These levels are retarded as reeulatorv minimum levels and to maintain these








Table


Illuminance Levels for Safety


Hazards Requiring Illuminance Levels for Normal Activity Level (in fc and Ix)

Visual Detection Low High
Slight 0.5 (5.4) 1 (11)

High 2 (22) 5 (54)

Source: Ref 8




Illuminance levels are also recommended to provide a guide for efficient visual

performance for various activities. Table 2.4 summarizes the recommended levels for

construction activities along with uniformity ratios, which are similar to the ones

recommended by CIE-Guide to the Lighting of Exterior Working Areas (42).




Table 2.4 Recommended Levels for Uniformity Ratio and Construction Activities


Activity


General Construction


Excavation Work


Illuminance Levels in fc (Ix)


10(108)


2 (22)


Uniformity Ratios


1 to 5


1 to 5


Source: Ref. 8


The recommendations of the IES are approved as American National Standards by


the ANSI Board of Standards.


workzone lighting: 1) RP-7-91


There are two ANSI standards relevant to construction


Industrial lighting, and 2) RP-8, Roadway lighting (43,


Standards on industrial lighting relate to industrial activities both indoor and outdoor and is


* a a a a ---


, r r r r r




35


lighting provides recommended maintained luminance and illuminance values and uniformity

ratios for various types of roadways.


OSHA Standards.


Occupational Safety and Health Administration (OSHA) has specified minimum


illumination intensities for construction and related areas.


These values, taken from ANSI


standards and IES recommendations, are included in OSHA Safety and Health Standards (29


CFR 1926/1910) and are presented in Table


For other areas or operations not


covered in the table, refer to the American National Standard Al 1.1


-1965, R1970, Practice


for Industrial Lighting, for recommended values of illumination.


Section 7


of the Code of Federal Regulations (CFR) number 30 provides detailed


illumination standards for the mining industry (46).


According to the code 0.06 foot lambert


of luminous intensity is required on the surfaces that are in a miner's normal field of vision.

For self propelled machines, the illuminated surface which is within 10 ft of the front and the


rear of the machine should have a luminous intensity of not less than 0.06 foot lamberts.


addition to illumination level in working places, the code also addresses the issues of lighting


fixtures, methods of measurement, and mining machines and cap lamps requirements.


Table


2.6 provides the proposed regulations for illumination requirements of various equipment


such as


- draglines, shovels and wheel excavators (47).


Figures


2.7(a) and 2.7(b) explain the


method of measurement of luminous by intensity direct measurement and by averaging

method, respectively (46).















Table


Minimum Illumination Intensities for Construction Industry


Area or Operation Illumination Intensity
Area or Operation n
(in fc)

General construction area lighting 5

General construction areas, concrete placement, exca- 3
vation and waste areas, access ways, active storage
areas, loading platforms, refueling, and field mainte-
nance areas.

Indoors: warehouses, corridors, hallways and exit 5
ways
5
Tunnels, shafts and general underground work areas:
(Exception: minimum of 10 foot-candles is required at
tunnel and shaft heading during drilling, mucking and
scaling.)
10
General construction plant and shops (e.g., batch
plants, screening plants, mechanical and electrical
equipment rooms, carpenter shops, rigging lofts and
active store rooms, barracks or living quarters, locker
or dressing rooms, mess halls, and indoor toilets and
work rooms).
30
First aid stations, infirmaries, and offices


Source: Ref. 45
















































a,


ECa
CL.I


0(0
b-V
w-x
w


r5
ca)- -


O

E
ZS 5


































a) Direct Measurement of Luminous Intensity


b) Luminous Intensity by


Averaging Method








State DOT's


Specifications


As evident from construction literature, nighttime work is common on many highway


projects across the country particularly in Florida and California.


showed that most of the highway agencies,


Questionnaire survey results


with a few exceptions, do not have detailed


specifications pertaining specifically to nighttime construction.


A summary of these results


by state are presented in


Table


The states having detailed lighting specifications


concerning nighttime construction include -

Michigan, North Carolina and Virginia. For v


only the minimum illumination intensities required (e.g.,


adequate for all construction activities.


California, Connecticut, Florida, Maryland,


iorkzone lighting such specifications mention


fc, etc.), which is presumed to be


In addition to this, some specifications require the


contractor to provide sufficient light for safety of personnel and quality of work.


most of the contractors develop their own systems.


As a result,


Considering project variations, and few


or no standards to go by, workzone lighting has little uniformity.

The FDOT standard specifications devotes a section 8-4.1 to nighttime construction


only.


It requires the contractor to furnish and place all lighting facilities on site and maintain


light intensity of


fc minimum at all times.


Lighting may be accomplished by the use of


portables, standard equipment lights, street lights, etc.


The contractor is also required to


submit a lighting plan at the Preconstruction Conference showing the type and location of


lights to be used (48). Similar specifications are included in the standards of Illinois and New

York highway agencies. In California, the minimum illumination intensities are established


* ~.1 *1n a.4 -a n n( n ar a









Table


Summary of Provisions for Lighting Requirements and Guidelines
for Various States


Name of the State Brief Description of the Provisions for Lighting

Connecticut Paver 6 fluorescent and 4 flood lights;
Roller 2 flood and 2 spot lights
Florida Min 5 fc
Maryland Min 20 fc (215 Ix); 30 ft (9 m) high tower; beam angle < 60 deg
Michigan Paver min 10 fc, 6 ft tower with 5 lights; 100-200 ft behind
Roller min 10 fc; 4 ft tower with 4 lights; 50-100 ft front & behind
Workzone min 10 fc
North Carolina Tower 30 ft high, lumen 50,000-460,000, min 20 fc, angle < 60 deg
Machine 13 ft high, lumen range 22,000-50,000, min 10 fc
Virginia Min 50 fc in 15ft x 15ft area and 5 fc in the corners







of 50 fc of light for areas of approximately 15 ft by 15 ft with a minimum of


fc in the


corners of the area (49).

The State of Connecticut has recently included a contract provision "Lighting for


Night Paving" in several nighttime paving projects.


The provision addresses lighting require-


ments for the paver and roller only to illuminate the work area and equipment.


contractor is responsible to furnish, mount and maintain all the required lighting, which is

inspected by an engineer for its adequacy prior to allowing a nighttime paving operation to


commence or continue.


Description of guideline for equipment is as follows:


PAVER-


Fixture Type


Ouantitv


Remarks


Fluorescent, twin 48" HO

Fluorescent, twin 48" HO


Mount over screed area

Mount over auger


PAR 38


15W Flood


Aim 25-50 ft behind paver


ROLLER-


Fixture Type


Quantity


Remarks


QPAR64 1000W Flood


QPAR64 1000W Spot


Mounted above roller

Aim floodlights 100 ft in front
of and behind roller. Aim


spotlights 200 ft in front of and
behind roller.

According to the provision, the contractor is responsible for providing portable diesel

generators on rollers and pavers to adequately furnish 120v electric power to operate the

qnpeifipd liohtino pnnlinment DPlion and fahrirntinn Anfhrac.kt and hardware should be done








and pavers.


The issues of glare to traffic, obstruction to traffic, safety codes and adjustable


mounting are also briefly mentioned in the contract provision.


The cost for providing any


lighting is not measured for payment separately.

A similar special provision is also included by Maryland State Highway Administration


for nighttime pavement repairs.


20 fc (21


According to the provision, if existing light levels are below


Ix) over the construction area, supplemental or flood lighting should be utilized.


The contractor is required to submit a situation plan showing the locations and aiming of the


floodlights.


The lighting system will be checked for proper aiming, positioning, level and


uniformity of illumination and any hazard to maintenance of traffic.


Other specifications as


outlined in the special provision include

Illumination level:not less than 30 ft (9 m) for directly influenced traveled roadway

Mounting height: not less than 20 ft (6 m) for indirectly influenced traveled roadway


No additional payment is made for the contractor utilizing nighttime construction.


the material and equipment for lighting is fUrnished by the contractor and remains the

contractor's property.

Michigan Department of Transportation also has included a special provision for


lighting for night paving.

general workzone lighting.


The provision outlines lighting arrangements for paver, roller and

The summary of nighttime paving operation and requirements as


addressed by the provision is as follows:




43


PAVERS-

illumination level on and around paver should be a minimum of 10 fc.

a tower with minimum height of 6 ft should be mounted on the paver holding five

lights.

one light on the tower should be facing forward and four lights facing toward the new

mat being laid.

area behind the paver should be lighted up to 100 ft to 200 ft.

ROLLERS-

each roller should be equipped with four headlights-two facing each direction of

travel.


a tower with a minimum height of 4 ft should be mounted on the roller equipped with


four lights.

two lights of the tower should face each direction of travel lighting the area 50 ft to

100 ft.

WORKZONE-

auxiliary lighting should be provided for the entire paving operation.

each light should be a minimum of 500-watt quartz.

ramps, gore areas or areas of extensive hand work should have separate supplemental

lighting of 500-watt quartz minimum.

The State of Michigan has also included a supplemental specification for nighttime


casting of superstructure concrete.


One section of the specification deals with the lighting and




44


State of North Carolina has included a section describing portable lighting which is

used for compliance with the standard specification section on night work requirement for


artificial lighting.


The department requires the contractor to present a lighting plan on


standard size roadway plan sheets. Such a plan must clearly show the location of all lights

necessary for every aspect of work to be done. The section deals with tower and machine


lights separately and can be summarized as below:

TOWER LIGHT-

height of tower light should be 30 ft.


light should consist of mercury vapor,


metal halide, high pressure or low pressure


sodium fixtures.

each fixture should have a minimum output of 50,000 lumens and combined output

of all fixtures should not exceed 460,000 lumens.

should provide an average maintained horizontal illuminance greater than 20 fc over

the work area.

main beam of the light should not be aimed higher than 60 degrees above straight

down.

MACHINE LIGHT-

mounted on supports attached to construction equipment at a height of approximately

13ft.

each fixture should have light output between 22,000 lumens and 50,000 lumens.

electrical grounding of generators to the frames of machines should be in accordance




45


these lights are in addition to conventional equipment headlights.

In addition to light specification there is a small section briefly addressing construction


methods and lighting arrangements.


Payment for the lighting provided by tower and


equipment lights is made at the contract lump sum price for "portable lighting."


Current Practices


Nighttime Work


In a study of construction practices related with nighttime pavement, at various

airports with asphaltic concrete, it was found that certain clauses dealing with construction


lighting were included in the contract specifications (50).


According to the specification, the


contractor was required to install, maintain and relocate temporary light to illuminate the


working areas during the hours of darkness when overlay operations are in progress.


lighting equipment was to be trailer-mounted units, each with four 1000-watt metal halide or


high pressure sodium lights on a winch-lift telescopic mast.


The contractor was to maintain


an average illumination level of 5 fc throughout the working area and should submit isolux


curves or charts showing the pattern of lights.

accordance with standards of IES current prac


Levels were to be calculated and measured in

tice. In addition, all paving machines, rollers,


distributor trucks and other equipment (except haul trucks) were to be equipped with artificial

illumination sufficient to safely complete the work.

Minimum illumination level was to be 5 fc and was to be maintained in the following


areas:








An area


ft wide and 12 ft long immediately behind the asphalt spreader during the


operation of the machine.

12 ft wide by 30 ft long area immediately in front and back of all rolling equipment.


12 ft wide and 12


ft long at any point where tack coat is being placed prior to place-


ment of the hot mix asphalt overlay.

12 ft wide by 20 ft long immediately in front of the cold milling machine.

12 ft by 30 ft long immediately in front of the heater scarifier and heater planing

machine.

12 ft wide by 30 ft long in front of the rubber asphalt distributor and spreading equip-

ment.


12 t wide by 12


ft long at any point where joint sealing operations are in progress.


This level of illumination can be obtained with four 1000-watt metal halide floodlights

at 30 ft mast height aimed at 60 degrees and placed at approximately 200 ft centers along

each edge of the runway (50).

An illumination case study of a stripping shovel by Faux suggested that an illumination

intensity level of 10 fc on the workface and around the machine was found satisfactory by the


operators for visual tasks (47).


In the study it was recommended that a 20 ft area on all sides


of the machine and 100 ft in the front should be illuminated with minimum 10 fc intensity.

To provide good color rendition a combination of metal halide as the high intensity discharge

(HID) source and tungsten halogen as the incandescent type was utilized and was found to

be providing a highly satisfactory quality of light for the application.







units were used as necessary for supplemental lighting.


For a paving operation in New Jersey,


auxiliary generators of 3500-kw low RPM 4 cycle were used to take care of additional


lighting (52).


On the same project, rollers were equipped with five 300-watt all weather bulb


clusters at the front and back; whereas, on the spreaders, due to vibration, four heavy duty


bulbs of400-watt sealed beam clusters were used. At asphalt plant site six 1500-watt wide

angle mercury vapor lights were used at strategic locations. These lights were sealed and had


highly polished reflectors.

According to one study on a project in Denver, Colorado, insufficient lighting was a


problem (53).


It was found that existing light, about 1 fc, produced from light standard is not


sufficient for construction procedures.


For crack filling,


500-watt spot lights mounted


approximately 10 ft high on each corner of the crack filling machine were used.


The work


area was measured to be 36 fc with light meter and appeared to be sufficiently lit for the


workers to work without any restraint.

requirements for similar tasks. The pay


This conformed to the IES minimum light levels


ing machine was equipped with five high intensity


lights, two on each side of the machine and one in the center, and the lighting was found to


be adequate. The roller used only one spotlight along with the headlights, which did not seem

to be adequate. Lighting systems attached to the equipment worked well when the crew


doing the hand work had control over the speed of the vehicle such as the crack filling

operation, however for paving because of its fixed speed such a system did not work well.

In a project in San Diego, California, six Allmand maxi-lites were placed on the perim-


eter of the section being paved (50).


The rollers had their own headlights to provide


-- "







obtained with four


1000-watt metal halide floodlights at 30 ft mast height aimed at 60


degrees.


Lighting Sources and Equipment


To obtain information concerning current construction workzone lighting practices,


various lighting manufacturers and suppliers were contacted.


Although most manufacturers


did not indicate substantial construction lighting experience, it was, however, noted that

incandescent and HID types of sources are most commonly used in construction workzones.

In Florida most of the construction sites were found to be equipped with four 1000 watt metal

halide light towers.

Incandescent lamps include tungsten halogen lamps and HID includes mercury vapor,


metal halide and high pressure sodium.


disadvantages.


Each lamp has its respective advantages and


Incandescent lamps have an advantage of lowest initial cost, instant on, no


ballasts, natural day light color, good light aiming control and very low lumen depreciation.

However, low lumen output and short lamp life are some of the drawbacks with incandescent

lamps in addition to their high surface temperature.

In the HID group, mercury vapor lamps also are vibration resistant and fast restrike


time.


They also have longer life; but, lower lumen output and high lumen depreciation are


some of the drawbacks.


Metal halide lamps, one of the most commonly used for flood


lighting purposes, are known for their good color rendition and high lumen output.


they have long restrike time and medium lamp life,


Although


they are good for overall lighting.







Other


advantages


of high


pressure


sodium


include:


lumen


output,


lumen


depreciation, reduced glare and total system cost.


Table


2.8 provides a summary of ratings


of various lamps used for construction lighting (8).


In general, HID lamps have longer lamp life and efficacy (lumens per watt).


a time delay and restrike time when the lamps are first started.


There is


Typically due to their high


efficacy, less luminaries are required than for an incandescent type of lighting.


In addition,


because of their rough service capability and better vibration handling, HID lamps are

considered suitable for highway workzone lighting.

Usually for highway construction lighting HID lamps are grouped in sets of two or


four lamps. These luminaries are typically mounted on portable trailers and are accompanied

by generators. The luminaries are commonly narrow beam and used for area lighting or flood

lighting. Incandescent lamps, however, are usually mounted on highway construction


equipment and used for spot lighting of the tasks.


Field Investigation


A number of field trips were made to investigate the current FDOT nighttime highway


construction and maintenance practices.

graphical location and diversity. In select


The selection of projects was based on their geo-


.ing projects in Florida not only was an effort made


to select representative kind of typical highway projects, but care was taken to avoid

repetition and a variety of projects covering nearly all the operations were chosen.

A field review form was prepared to record all relevant information as observed on













C

Si
o4-
'4-


.5


SCr
00

04.


a



* -
6.


-C4




10


a)
0


(U


Ca>L
L. caE
0o


a

LL -
9% u





E
-J
.

.1


a-
s o

Z r


ONY


Owm


E L
-J







of seeing; general lighting information, lighting equipment, and configuration; quantity of light

and illumination levels: uniformity, direction and glare of light; and general information about

workers and power sources.

During field reviews several projects with different nighttime operations were visited.

This included projects in rural, urban and semi-urban environment on limited access, primary


and other types of roadways.


Various types of operations included: replacing concrete barrier


to separate traffic from workzone; asphalt concrete paving of intersection; in-situ


concrete construction of bridge deck; excavation, filling and embankment construction; and


milling, repaving and marking of limited-access highway.

observation forms are included in Appendix A.


A list of FDOT projects and field


Various tasks were identified related to all the observed operations.


Other infor-


nation such as-background reflectance, importance and speed required, and seeing distance


for the tasks were collected.


To provide common lighting visually most difficult and fatiguing


tasks were also identified in each operation.


Most of the observed tasks had low or medium


background reflectance owing to low reflectivity of the pavement and other concrete and soil


structures.


Importance, speed and accuracy of the tasks also varied from low for excavation


to high for paving and finishing works.


Seeing distances of the tasks were categorized into


four main categories: 1) less than one ft, 2) one to five ft, 3) five to 1


ft, and 4) more than


For construction lighting, equipment mounted lights, portable light plants and their


combinations were most commonly used.


Some equipment such as pavers and rollers were







illuminating the sides of the equipment.


Compacting rollers and brushing rollers usually had


two to four sealed lamps and sometimes additional mounted lights.


However, wheel loaders,


dump trucks and flat-beds were equipped with manufactured conventional equipment lights.

For bridge deck construction at Fort Myers, FL, the crane, conveyor system and screed were


equipped with several mounted lights. Mi

manufactured and installed at the factory.


lling machine had sealed beam units which were

Relatively stationary workzones were usually


supplemented by portable light plants which were also used for general lighting of the area.

Most of the mounted lights were 500-watt tungsten-halogen lamps powered by diesel


generators.


The generators, in most areas,


were installed on the construction equipment.


These halogen lights were mounted on custom made brackets and poles and provided


sufficient flexibility to change lamp height and aim as desired.


Most of the fixed lights were


to 100-watt conventional automobile lamps and had fixed aim and light positions.


most common portable light found during field reviews was light trailer equipped with four


HID lamps of 1000 watts each. Trailers were also equipped with diesel generators and 30

ft maximum height adjustable tower. These light plants utilized metal-halide as one of the


most common lamps enclosed in a parabolic reflective cover to provide a uniform narrow


beam of light.


These light towers provided sufficient uniform flood light for the workzone,


however, in some instances they also caused severe discomfort glare and sometimes disability








glare to the motorists especially when installed against the moving traffic.


For the bridge deck


construction project, the concrete barge and chute were illuminated with 1000-watt sodium


vapor lamps enclosed in a rectangular reflective covering.


Lights on the crane were 1000-


watt metal halide lamps spaced at 30 ft and oriented towards the hoist.


For the milling-


repaving project on 1-295, Jacksonville, FL, lights on the milling machine were fixed and

oriented to illuminate the critical areas such as the conveyor, milling edge, rear of the

equipment and on the sides.

Quantity of light was found to be sufficient for most of the tasks, however, some tasks

were not adequately illuminated; reasons for which were attributed partly to the inadequacy


of light plants and partly to their improper orientation.


Particularly for compacting rollers,


lighting was not sufficient and the operator was moving the equipment broadly in a pattern

based on the experience instead of visual task. Similarly illuminance levels were not enough


for sweeping brush roller and asphalt spreader applying tack coat.


The operators essentially


moved in a certain predefined pattern and sometimes failed to notice missed spots.


intersection paving and bridge construction jobs also, illuminance levels for hand spreading


of mix were less than satisfactory.


In most cases, illumination of the general area was found


to be adequate, however, in many cases task illumination was not adequately emphasized.

Factors which were observed to evaluate quality of light included uniformity, direc-


tion, diffusion, direct and veiling glare.


adequately uniform.


Lighting for nearly all the observed operations was


In milling and repaving operation, uniformity of light was difficult to


maintain because of the continuous movement of the operation.


Good uniformity was





54


more spot and task illumination. Equipment lighting was usually factory installed sealed-beam

units.


The direction of lighting was found objectionable in many cases, particularly for tower


lighting plants.


For replacing barrier walls on 1-75, Gainesville, FL, the plant was placed in


close proximity of the travel lane facing oncoming traffic, the result of which was nearly


blinding disability glare to the motorists.


In other cases, light plants were situated at locations


creating a shadow zone on the tasks. Because of inappropriate directivity of lights, on several


occasions tasks were performed in negative contrast instead of a positive one.


defects aided their identification.


Shadows of


Spotlights mounted on various equipment, in general, had


better directivity. Lights on some of the compaction rollers were not mounted high enough,


as a result most of the light fell on the wheels instead of the pavement.


Spotlights on milling


machine were factory installed and had better directivity.

Veiling glare was negligible for all the observed operations because of the low reflec-


tivity of pavement and other construction surfaces.


Direct glare to workers was, in general,


more common in the case of highway operations than for bridge construction.


The problem


of glare to motorists was found to be acute for highway operations in which adjacent lanes


were opened to traffic.


In urban and semi-urban environment, particularly where roadway


lights were available, problems of glare were less, due to reduced background contrast.


was noted during field investigation that lighting design and provision was essentially based

on the contractor's discretion, and in some cases, little or no thought was given to location,

positioning and orientation of the light plants.








Summary


In this


chapter


an overview


of the


state-of-the-art


of lighting


nighttime


construction practices was presented.


For organization the chapter was structured in three


parts each of them dealing with review of literature, lighting standards, and current practices


respectively.


Review of literature consisted of extensive survey of basic concepts in lighting,


studies pertaining to design procedures, and effect of human factors in determining adequate


lighting.


The important parameters of visual task were evaluated including illuminance,


contrast, and reflectance.


Discussion on human factors consisted effect of glare sensitivity,


transient adaptation,


and other physical and psychological factors on safety,


accidents,


productivity and performance.


In order to develop useful and consistent guidelines in this study,


guidelines and standards were discussed.


various existing


Among these standards the prominent ones were


from IES, OSHA, and specification of various state highway agencies.


Current nighttime


highway construction practices were also reviewed through field investigation and survey


questionnaire.


Lighting manufacturers were also contacted to provide information on the


presently used lighting systems for


construction lighting.


In the remaining part of the report


model components are discussed and developed followed by development of guidelines.












CHAPTER 3
MODEL COMPONENTS AND THEIR FORMULATION


Introduction


To formulate the model,


the model components needed to


be determined and


described.


The fundamental questions were: 1) Which activities will be included in the model


and for guidelines? 2) Which


independent and dependent factors will be chosen for the


model? and 3) How the preliminary levels for highway tasks will be determined?


To answer


these questions, this chapter


has been divided into three parts.


Each part separately deals


with the identification of typical highway construction activities, determination of factors

affecting illumination requirements for a particular task, and development of a non-highway

matrix for comparative analysis.

Nighttime construction activities were identified with the help of literature and survey


of various state highway agencies.


A questionnaire was prepared and sent to agencies


nationwide and responses were compiled and analyzed to prepare a list of typical tasks


commonly performed at night.


In the second part of this chapter from the analysis of


literature review, a number of factors were identified.


A list of five significant factors were


selected from the above list to be used as independent parameters in the model.


comparison with the existing industry standards, a list of non-highway tasks was identified.







Nighttime Work Activities


Preliminary Identification


In order to categorize typical highway nighttime work, various highway operations


were identified.


Interviews with state highway personnel, opinions of various knowledgeable


individuals and review of FDOT standard specifications for road and bridge construction

resulted in a preliminary list of the most commonly performed highway operations (48).

These operations were categorized into highway maintenance and highway construction tasks.


Both lists also included activities on bridges, signalization and other highway facilities.


brief description of these tasks is presented in Table 3.1.


The tasks in each list represent


various operations and activities which are categorized according to their similarities in visual

requirements.

Table 3.2 shows some of these task categories, typical operations represented by

them, and various activities involved in the operation. Although all the activities in a particular

operation (as shown in Table 3.2) may not have similar visual requirements, they are grouped

together for practical reasons. Compliance is more realistic if a single lighting standard is

specified for one operation rather than a different standard for each activity in the operation.

However, the task category representing several operations is based on the similarity of visual

requirements of those operations.




































0)
E
a)


C -r


r: i cj t: c


r: i :ui cb rd i6d
i- ,- ,- r r rC


~: ed dO
r









Table


Categorization of Typical Highway Construction Tasks and Operation


Task Category Type of Task Activities

Excavation Regular excavation
Subsoil excavation
Lateral Ditch excavation
Channel excavation
Embankment Dry fill method
Hydraulic method
Backfilling Pipe Culverts
Storm Sewers
Other Structures
Subgrade Stabilization of subbase
Lime-treated
Cement-treated
Base Courses Limerock base composition of mixes
Graded aggregate base preparation of subgrade
Sand-Clay base spreading the mix
Shell stabilized base compacting and finishing
Soil-cement base correcting defects & thickness
l____ Asphalt base priming, curing or maintaining
Surface treatment Prime coat preparation of surface
Tack coat distribution of material
spreading cover material
rolling and curing
Cement-concrete Pavement subgrade preparation
Sidewalk setting forms
placing reinforcement
placing concrete
consolidating and finishing
straightedging & surface correction
joints and curing
Fencing Guardrail setting timber & steel posts
Fencing placing rail or fabric
reflector elements
Highway lighting excavation and backfilling
trenches for cable
concrete base for light poles
,.CCA~ fk-Aj ^ n iac 1^r rlt>








Summary of Survey Ouestionnaire


A survey questionnaire including the preliminary list of maintenance and construction


tasks as


identified above was sent to all the state highway agencies.


A copy of the


questionnaire survey containing these lists and other questions is provided in Appendix A.

In the questionnaire, respondents were asked to identify the maintenance and construction


tasks performed in their states during nighttime.


frequency of these nighttime tasks.


in Table 3.3.


They were also asked to indicate the


The results of the questionnaire survey are summarized


A similar survey of all district offices of FDOT resulted in a number of projects


performed at night. A summary of the survey results is provided in Table 3.4.


Out of


a total of 33 states responded, of which 28 respondents indicated some


nighttime construction and maintenance work in their states.


Some state highway agencies


including California, Connecticut, Florida, Illinois, Maryland, Michigan, Missouri, New York,


North Carolina,


Virginia, and Washington indicated a considerable amount of nighttime


highway work in their states-usually 10 percent or higher


of all the awarded projects-while


other states such as Arkansas, Mississippi, New Mexico, North Dakota and South Dakota

reported having no experience with nighttime highway rehabilitation work.

In addition to quantity of night work, Table 3.3 also addresses the issues of applica-


tion of screens or barriers between the travel lane and workzone to avoid


glare to motorists


and availability of lighting standards or specifications to determine workzone lighting.












Table


Summary of Questionnaire Survey of State Highway Agencies


Name of the State Number of Nighttime projects Use of Screen Lighting
performed in a year or Barrier Standards
Maintenance Construction


Arkansas
California
Colorado
Connecticut


Dist. of
Florida


Columbia


Hawaii
Idaho
Illinois
Iowa
Kansas
Kentucky
Maine
Maryland
Michigan
Mississippi
Missouri
Nevada
New Jersey
New Mexico
New York
North Carolina
North Dakota
Oklahoma
Pennsylvania
Rhode Island
South Dakota
Tennessee
Texas
Utah
Virginia
WAchinntnn


0
80
15
28
0
10
-?
0
0
1-10%
2
0


Lu JL .- I --








Table 3.4


Summary of Questionnaire Survey of FDOT District Offices


District

No.


1
2
3
4
5
6
7
Turnpike


I I


District

Office


Bartow
Lake City
Chipley
Ft. Lauderdale
Deland
Miami
Tampa
Plantation


Number of Nighttime Projects

performed in a year


Maintenance


'?' indicates that the number of projects is undeterminable


Compilation & Analysi


Construction


of Responses


The results of responses obtained from the nation-wide survey regarding construction


and maintenance tasks during nighttime are summarized in Tables 3


and 3.6, respectively.


Tables 3.5 and 3.6 give the relative frequency of night work as reported by various state high-


way agencies.


To obtain most commonly performed maintenance and construction tasks,


both the lists are sorted and arranged in a decreasing order of the state frequencies.


Similar


lists are prepared for FDOT nighttime highway construction and maintenance tasks based on


the information obtained from all the seven district offices.


The lists are provided as Tables


and 3.8 for construction and maintenance tasks, respectively.


These lists are also


arranged in the decreasing order of their frequencies.













Table 3.5


Number of States Performing Various Nighttime Highway Construction Tasks


Task Construction Tasks performed at night F___requency of Tasks
No. Frequently Occasionally Rarely
1. Resurfacing 7 8 12
2. Barrier Walls, Traffic Separators 7 8 3
3. Milling and Removal 6 11 9
4. Painting Stripes, Pavement Markers, Metal Buttons 4 8 4
5. Bridge Decks construction 3 16 7
6. Concrete Pavement construction 3 9 12
7. Base Courses clay, cement, asphalt 3 6 8
8. Excavation regular, lateral ditch, channel 3 5 10
9. Embankment, Filling and Compaction 2 6 10
10. Highway Signing for construction 2 6 8
11. Subgrade stabilization & construction 2 5 10
12. Surface treatment 2 1 5
13. Drainage Structures, culverts & sewers construction 1 5 13
14. Waterproofing/ Sealing 1 4 6
15. Construction of other Concrete Structures 0 6 13
16. Guardrail, Fencing 0 4 5
17. Highway Lighting System construction 0 3 7
18. Traffic Signals construction 0 1 9
19. Landscaping, Grassing, Sodding 0 0 9
20. Riprap placement 0 0 6
21. Sidewalks construction 0 0 4
Total number of responses 28


Note


: Tasks are arranged in the decreasing order of their frequencies.

















Table 3.6


Number of States Performing Various Nighttime Highway Maintenance Tasks


Task Maintenance Tasks performed at night Frequency of Tasks
No. Frequently Occasionally Rarely
1. Milling and Removal 5 3 0
2. Resurfacing 5 3 0
3. Painting Stripes/ Pavement Markers 4 3 1
4. Sweeping and Cleanup 3 1 2
5. Barrier Wall or Traffic Separator 3 1 1
6. Crack filling 2 0 3
7. Repair of Concrete Pavement 2 0 2
8. Bridge Decks rehabilitation & maintenance 1 5 0
9. Reworking Shoulders 1 3 2
10. Drainage Structures maintenance & rehabilitation 1 3 0
11. Pot filling 1 1 4
12. Highway Signing for maintenance works 1 1 3
13. Traffic Signals maintenance 1 1 1
14. Landscaping/ Grassing/ Sodding 1 1 0
15. Resetting Guardrail/ fencing 1 1 0
16. Highway Lighting System repair & maintenance 0 2 3
17. Waterproofing/ Sealing 0 2 3
18. Maintenance of earthwork/ embankment 0 2 2
19. Riprap maintenance 0 1 0
20. Surface treatment 0 0 3
21. Sidewalks repair & maintenance 0 0 1
Total number of responses 9


Note: Tasks are arranged


n the decreasing order of their frequencies.
















Table 3.7


Performing Frequency of Various Construction Tasks on FDOT Nighttime
Projects


Task Construction Tasks performed at night Frequency of Tasks
No. ____Frequently Occasionally Rarely
1. Construction of other Concrete Structures 5 2 1
2. Painting Stripes, Pavement Markers, Metal Buttons 4 5 0
3. Resurfacing 2 1 3
4. Milling and Removal 2 1 2
5. Subgrade stabilization & construction 2 1 1
6. Bridge Decks construction 1 7 1
7. Base Courses clay, cement, asphalt 1 2 1
8. Barrier Walls, Traffic Separators 1 1 3
9. Guardrail, Fencing 1 1 3
10. Waterproofing/ Sealing 1 1 2
11. Highway Lighting System construction 0 5 2
12. Concrete Pavement construction 0 5 1
13. Highway Signing for construction 0 4 3
14. Drainage Structures, culverts & sewers construction 0 2 3
15. Excavation regular, lateral ditch, channel 0 1 3
16. Traffic Signals construction 0 1 2
17. Riprap placement 0 1 1
18. Surface treatment 0 1 1
19. Landscaping, Grassing, Sodding 0 1 1
20. Embankment, Filling and Compaction 0 1 0
21. Sidewalks construction 0 0 2
Total number of responses 10


Note: Tasks are arranged in the decreasing order of their frequencies.













Table 3.8


Performing Frequency of Various Maintenance Tasks on FDOT
Nighttime Projects


Task Maintenance Tasks performed at night Frequency of Tasks
No. Frequently Occasionally Rarely
1. Sweeping and Cleanup 7 8 3
2. Repair of Concrete Pavement 6 2 12
3. Bridge Decks rehabilitation & maintenance 4 4 8
4. Resurfacing 3 7 9
5. Milling and Removal 2 8 7
6. Highway Lighting System repair & maintenance 2 8 4
7. Traffic Signals maintenance 2 7 7
8. Painting Stripes/ Pavement Markers 2 5 7
9. Surface treatment 2 1 2
10. Barrier Wall or Traffic Separator 1 4 8
11. Crack filling 1 3 6
12. Pot filling 1 3 5
13. Resetting Guardrail/fencing 1 3 3
14. Waterproofing/ Sealing 1 2 3
15. Highway Signing for maintenance works 0 3 11
16. Drainage Structures maintenance & rehabilitation 0 3 3
17. Sidewalks repair & maintenance 0 2 3
18. Reworking Shoulders 0 1 6
19. Riprap maintenance 0 1 5
20. Landscaping/ Grassing/ Sodding 0 1 3
21. Maintenance of earthwork/ embankment 0 0 7
Total number of responses 27


Note


: Tasks are arranged in the decreasing order of their frequencies.





67


hand, traffic signalling, lighting systems, landscaping, rip rap and sidewalks are the least


preferred tasks for nighttime construction.


Table 3.6 shows cleanup, concrete pavement


repair and bridge deck rehabilitation as frequent maintenance activities during nighttime.


the survey, many respondents indicated that a significant amount of nighttime maintenance


is attributed to emergency work in addition to regular required maintenance.


Earthwork,


landscaping and rip rap are the less common maintenance work conducted at night.


Factors Influencing Illumination Reauirements


From literature review and discussion with experts on lighting and illumination, a


number of factors affecting illumination requirements were identified.


During the process of


identification only those factors which are related with outdoors and nighttime highway type


situations were selected.


They are categorized in four categories, which include: 1) human


factors, 2) environmental factors, 3) task-related factors, and 4) lighting factors.

classification was done for the task-related factors due to their varying characteristics.


3.1 shows a summary of all the factors and categories.


Further

Figure


These categories are explained in


greater detail in following sections.


Human and Cognitive Factors


Visibility detection and recognition are greatly affected by various human physical,


physiological and psychological factors.


The important human factors affecting vision and


perception are: 1) age, 2) visual acuity, 3) response characteristics, attention, and expectation,





















rn


C


)IL

OL
RI


C

Co



ml


-o





I~g


1


r


i


'C

0)


0...


em
I-


Pwe


C10
Cc
-Jo


a


ra.




RI




69


Ability to detect targets tends to decrease for the observers belonging to a higher age group.


Similarly,


visual acuity which is influenced by several human visual functions such as glare


sensitivity and transient adaptation, affects an individual's perception and recognition.

On the other hand, response characteristics and cognitive behavior of an individual,

experience and familiarity with a particular object also affect the process of recognizing and


perceiving that object.


For example, an experienced operator can identify defects on a newly


paved road surface even under reduced illumination levels as compared to one having less

experience with the work environment.


Environmental Factors


In this category, factors such as weather conditions, fog, surface condition and


ambient brightness are included.


As has been reviewed in the literature, ambient brightness


helps reduce the relative contrast of the object and the background which is crucial in


detecting an object.


However, for night work conditions, ambient brightness helps increase


the illumination level of the task and requires less additional light for adequate illumination.

Atmospheric losses may have a significant effect upon the illuminance at the observer's


eye unless the viewing distance is short.

at relatively short distances. Since for n


In fog, the law of diminishing returns takes effect


early all the highway construction and maintenance


work, distances are relatively short and visual field is more or less limited to the workzone,

weather conditions and atmospheric losses may not contribute significantly to the losses in

object visibility.








Task-Related Factors


Task-related factors, owing to their characteristics and variations, are further sub-cate-

gorized in the following five categories.

Equipment attributes. Utilization of various construction equipment adds another


dimension to nighttime highway work compared to other external night work situations.


Fac-


tors related to construction equipment influencing task illumination requirements include: 1)


speed, 2) physical characteristics, and 3) response time.


exposure time or the duration of seeing.


Speed of equipment governs the


To some extent higher speed of a particular piece


of equipment can be compensated by increased illumination for adequate visibility.


Physical


characteristics of some equipment may restrict the vision of the operator and some of the


targets may not be detectable irrespective of the quantity of illumination. For example in

roller, operator's vision is restricted by the presence of drums in the visual field. Similarly for


loader, paver and other equipment also, detection of targets near the equipment is determined


by its position.


distance.

equipment.


However, for far targets quantity of illumination increases the detection


Required detection distance also depends on the response characteristics of the

Equipment with low response time, low maneuverability or high speeds require


greater detection distance and hence increased illumination.

Physical attributes of the task. Physical characteristics of the task or the target have


significant effect on the quantity and quality of illumination required for its detection.


Size,


type, appearance, and reflectance targets and their relationship with luminance values for their








target also helps its identification.


For the same illumination level, detection of any target in


the middle of the workzone is faster than the ones on the sides.

of a target from the observer also affects its identification. A d


likely to


Minimum distance of seeing


defect on road surface is more


be detected by a ground crew than an operator on the equipment for same


illumination level.

Task qualitative attributes. The qualitative factors associated with nighttime highway

construction tasks include: 1) importance of the task, 2) accuracy desired 3) visual difficulty


in performing the task, and 4) visual fatigue as a result of performance.


As has been reviewed


in the literature, performance increases with the level of illumination until an optimum level.


Beyond this point, further increases do not necessarily deteriorate performance.


From the


conclusion of previous studies it can be inferred that visually difficult and visually fatiguing


tasks may be performed better with an increase in the illumination level.


Similarly, for the


tasks requiring higher accuracy and additional attention, increased illumination is desired.

Background factors. Background factors include the characteristics of the surface on


which a task needs to be performed or a target needs to be detected.


Surface brightness,


which also depends on background illumination, reduces the positive relative contrast for the


Particularly when the degrees of reflectivity of the background surface and the task are


in the same range, detection and recognition becomes difficult.

Operation attributes. These factors are essentially associated with the type of highway


construction operation and the location where it is performed.


These factors include: 1) type


offacility, 2) facility environment, 3) traffic control, and 4) location on the highway.


These








ways, other arterials, collectors or local roads.


Any construction or maintenance operation


on limited access highways requires increased attention in forms of quality, safety and traffic


control.


Glare problem due to excessive illumination is more severe on these highways as


speeds of vehicles are higher and drivers do not expect any interference.


Facility environment


includes the geographical area of the operation such as urban, semi-urban, or rural.


For urban


areas, street lights and roadway lighting may provide sufficient illumination to perform


common highway construction tasks. The pr

because of considerable ambient brightness.


oblem of glare is also limited in the urban areas

However, for rural areas adaptation and glare


is a serious problem because of the sudden shift from darkness to a brightly lit environment

and then back to darkness, for a motorist passing through the workzone.

Illumination requirements for highway operations is also affected by the traffic control


plan designed for the operation.


Complete road closure or detour for a bridge or highway


construction may eliminate the problem of glare to the motorists.


Similarly, traffic control


plans may depend on the duration of the project. For example, a short-term project may need


different types of lights and illumination levels than a long-term project.


area on the highway or bridge, such as


Location of the work


- on the right-of-way, median, shoulder, ramp, or


intersection, also affects the illumination specification and requirements.


Since reflectivity of


the background surface depends on the location on the highway, illuminance values to


facilitate detection of a target may vary with the location of task.


For example, as shown in


Figure 2.1 the retro-reflectivity of grass, gravel shoulder, asphalt pavement, and concrete


pavement and deck is different and varies with the distance from the lamp.


As a result, illumi-





73


Lighting Factors


Factors associated with lighting provisions which can influence task illumination

requirements include: 1) geometric relationship, 2) orientation, 3) power and type of lamps,


4) gradient uniformity, and 4) glare.


Geometric relationships between light plant, observer


and location of the object sometimes effect the amount of illumination required as the

reflectance of background surfaces varies with difference in positions and hence the lumi-


nance.


Similarly, orientation, power and types of lamps also affect reflectance and contrast.


Uniformity is associated with the quality of illumination which is important for seeing detail


in the work area.


Uniformity, depending on the task characteristics, also affects the quantity


of illumination required to perform the task.

is a direct result of the quantity of illumination


Glare of both types, discomfort and disability,

i. Glare is not only affected by the power and


type of lamps but also by the size of light source, orientation and reflectance of the back-

ground surface.


Identification of Significant Factors


Although all the factors discussed above influence task illumination requirements, the


degree of influence varies.


A list of factors which significantly affect the illuminance levels


for highway construction tasks is given in Table 3.9.


The factors in the table have been


selected fiom the factors described above and as a result of literature review for specifically


for highway construction related visual tasks.


Various levels are assigned to the factors for


A


A I n.( I I 1i S. A P .









has been no priority or significance level associated with these factors.


Table 3.9


List of Factors Significantly Affecting Nighttime Highway Task Visibility


Name of the Factor


of the objects to be seen


Shape of the objects to be seen

Contrast of object & background

Age of Workers

Reflectance of the surface

Time spent in seeing

Importance of Task

Importance of Speed

Importance of Accuracy


Visual Difficulty


in seeing


Visual fatigue experienced

Seeing distance of the object

Safety & Glare considerations




Importance of Uniformity


I


Suggested Levels of the Factors


Very Fine

Flat

< 30%

under 40

< 0.3

few seconds

low

low

low

low

low


-Sft


Non-critical




Non-critical


Small

1 -dimensional

30-70%

40-55

0.3 0.7


-5 mmn


medium

medium

medium

medium

medium


15ft


Moderately


Important

Moderately


Important


Medium

3-dimensional

> 70%


over


> 0.7

half hr +

high

high

high

high

high

> 15ft

Critical




Critical


A '. ,L, 2 ---- II .L L,,1 L C.. t,, ,, .t I.. LI 1.. 1 --- --- -zC a aI *


Large


I I I I


__











requirements identically and vary in their degree of influence.


Some factors are envisaged to


have direct and consistent effect on the lighting levels required for seeing.


These conclusions


are drawn from the findings from literature review and experienced personnel.


As a result


another short listed summary of factors is prepared to further evaluate their effect and are


eventually included in the data analysis and model development.


The factors which are


included in the model for their significance are as follows


speed associated with the task


accuracy or importance desired for the task


reflectance of the background surface


seeing distance of the object from the observer

relative size of object to be seen.

None of the human factors are included in the list because of the presumption that


every crew or operator is a normal attentive observer in a given age group.


Moreover, during


field investigation, it was determined that on FDOT projects, in general, equipment operators


belong to one age group.


Visual and other variations among crew members for highway


construction work do exist. However, lighting guidelines should be based upon normal visual


capacity with sufficient allowance for individual variation.


Similarly, environmental factors


are also found to have an insignificant effect compared to the five selected factors, which are

all task related factors and are representative of their sub-categories.


Determination of Factor Levels


--





76


actual field conditions, certain subjective levels were assigned to these factors from practical


considerations.


Table 3.10 provides the factors and their subjective levels.


Non-Highway Task Matrix


Description


Although subject of lighting has not been studied in detail for tasks related to

construction industry, for other industrial applications there has been extensive work existing


in the area of lig

and standards.


hting.


A need was felt to take advantage of the already existing guidelines


As a result an approach is adopted to develop some sort of comparison


criteria and relate construction tasks to visually similar other industry tasks.


Other industries


excluding highway construction industry are addressed as Non-highway Industries.


Since the


non-highway tasks are distinctly different from highway tasks, developing a suitable and


reliable criteria was crucial to comparative analysis.

section were utilized to perform the comparison.


A set of factors identified in the previous

Care was taken to choose mostly outdoor


non-highway tasks as highway construction is an outdoor activity.


Sources of Information


The visual requirements in a typical highway construction task are similar to the visual


requirements of certain outdoor industrial tasks.


For comparison with highway tasks pur-


poses, a list of equivalent non-highway tasks was identified and their illuminance levels were















Table 3.10


Factors Influencing Task Illumination Requirements and Their


Subjective Levels


Task No. Factors Subjective Levels
1 Importance and accuracy of the task L Low
M Medium
H -High

2 Background Reflection L Low
M Medium
H -High

3 Speed N Not applicable
L Low
M Medium
H High

4 Size of the object to be seen F Fine
S Small
M Medium
L Large

5 Distance of Seeing S 1 to 5 ft
M -5 to 15 ft
L ->15 ft








Industry, 3) Petrochemical Industry,


4) Pulp & Paper Industry, 5) Industrial Outdoor Spaces


etc.


Development of Matrix and SAS Dataset


Based on the characteristics of the tasks identified for non-highway activities, they


were assigned different levels of various factors as determined in the earlier section.


Table


provides


matrix


selected


non-highway


outdoor tasks,


their


area,


factor


descriptions, and recommended illuminance levels.


the matrix provided in Table 3.11


A SAS dataset has been developed from


. The factors are assigned identification for SAS dataset,


which are: 1) F1 for Importance of factors, 2) F2 for Background reflection, 3) F3 for Speed,


4) F4 for Relative size,

illuminance levels. The


) F5 for Distance for seeing, and 6) LEVEL for suggested


dataset also includes fields for area and activity for non-highway


tasks.


Summary


In this chapter various model components were determined and formulated.


Typical


highway nightwork activities were identified from various department of transportation


documents.


This preliminary list was sent to all the state highway agencies and their


responses were summarized.

activities were modified. Ta


Results of the survey were incorporated and list of highway


sks involving similar activities were classified in the categories


for simplication of highway task list.








Table 3.11


Factor Description and Illuminance Levels of Outdoor Industrial


Tasks and Spaces


Area Name of the Activity F__actors __Recom.
_Imp. Refl. Spd. Size Dist. Level (fc)
Automotive Frame Assembly H M L S S 50
Industry Welding Area H H N S S 50
Machining Operations H H H F S 75
Coal Yards & Oil Storage L L N L L 0.5
Outdoor Substation, Parkin L L N L L 1.5
Entrance, Truck Maneuveri L L L L L 5
Furnace Area. Sheet Rollin H M N M S 30
Iron & Steel Mold Yard L L N M L 5
Industry Scrap Stock Yard M M N M L 10
Hot Top Storage M H M M L 10
Petrochemical Pump rows, valves, manifol L L N M L 5
Industry Heat Exchangers L L N M L 3
Maintenance Platforms L L N L L 1
Operating Platforms L L N M L 5
Cooling towers, equipment L L N M L 5
Furnaces L L N M L 3
Active Ladder, Stairs L L L L L 5
General Area L L N L L 1
Extruders & Mixers M M L M M 20
Conveyors L L M L L 2
Outdoor plants, equipment L L N S M 5
Outdoor Substation L L N L M 2
Plant Road: Freq. use L L M L L 0.4
Plant Road: Infreq. use L L L L L 0.2
Plant Parking Lots L L L L L 0.1
Outdoor bulk storage L L N L L 0.5
Large Bin Storage L L N L M 5
Small Bin Storage M M N S M 10
Small Parts Storage M M N F M 20
Pulp & Paper Groundwood Mill Grinder H M H S M 70
Industry Beater Room H M L M M 30
Roll Dryer H M M S M 50
Cutting & Sorting H M H S S 70
Active Warehouse M M L M M 20
Shipping Truck Shed M M L M M 20
Roadways L L M L L 0.4
Log Pile L L N M L 3
Inn I nlnndinn L L L M M 5





80


lighting requirements included: 1) Human & cognitive factors, 2) Environmental factors, 3)


Lighting factors, and 4) Task related factors.


From the list of factors, some factors were


short listed which had significant effect on the illumination.


Factor levels were also


determined subjectively to include in the model and to facilitate analysis.


A list of non-


highway tasks were selected from literature and their suggested levels were utilized to prepare


a SAS dataset.


This dataset is used for analysis in the next chapter.












CHAPTER 4
MODEL DEVELOPMENT


Introduction


It has already been hypothesized and also proven by a number of studies that the


selected factors such as accuracy and importance, reflectivity,


speed, size, and seeing distance


have varying influences on the requirement of an illumination level for a particular task.


order to determine a functional relationship between illumination level and various factors,


a model approach is adopted.


Although the significant affects of these factors are also


documented by some studies, their exact contributions are not known particularly in a


highway workzone type of situation.


Moreover, prediction of a definite lighting level for the


given conditions of a construction task is difficult in the absence of a quantitative approach.

In this chapter a mathematical model approach has been suggested for determining the


illumination level for any given highway construction task.

of model development procedure adopted in this chapter. 1


Figure 4.1 shows the flow chart


'he components formulated in the


previous chapter have been utilized to develop a problem of one dependent variable and five


independent variables.


A statistical general linear model is proposed and discussed.


simplify computing procedures, assistance of the SAS system has been solicited.


Various


relevant SAS procedures along with their applicability have also been described.




















DATABASE
DEVELOPMENT


CORRELATION
COEFFICIENTS


TRIAL MODEL


ELECTION


NEED TO ADD
MORE TERMS


APPLICATION OF
MODEL & TESTING


Figure 4.1


Flow Chart of Model Develooment Procedure


SELECTION OF PARAMETERS
1. Literature Review
2. Study of Work Process
3. Opinions of Experts


REGRESSION ANALYSIS

1. Least Square Fit


MODEL ADEQUACY

1. Residual Analysis for
fitted values
regressor variable
omitted variable
2. Normality of Residuals
3. Goodness of fit test
4. Coeff. of Determination


--




83


numerical values for non-highway industrial outdoor applications and their illumination


recommendations.

illumination level.


Data is analyzed to determine the correlation between various factors and

Several trial models have been selected and analyzed using SAS system.


Regression analysis, which is frequently used to analyze data from unplanned experiments,


is performed on all the models and the corresponding hypothesis has been tested.


evaluated for lack of fit and adequacy.


Models are


With the help of residual plots, regression coefficients


and error sum of squares, the best suited model is selected.


Model Approach


Modeling is defined as a process for determining, defining and explaining the relation-


ships among a set of variables.


Models usually involve two groups of variables.


They are


Dependent variable, sometimes also referred to as response variable, is defined as the

variable to be predicted from a given set of variables.

Independent variables, also sometimes referred to as predictor variables, are the other

variables that are to be evaluated in a research.

In order to predict the value of dependent variable, levels or settings of the inde-


pendent variables need to be taken into account.


The estimation problem can be greatly


simplified if models relating a response (dependent variable) to a set of independent variables


are considered.


To demonstrate these models, in this section, various linear and multiple


regression models are discussed


In the first part of this section a model is formulated and


estimation and test procedures are developed when the dependent variable is related to one




84


Regression Models


Suppose a single dependent variable or response y depends on independent or


regressor variable


The relationship


between


these


variables


is characterized


mathematical model called a regression equation.


The regression model is fit to sample data.


In some instances, the experimenter knows the exact form of the true functional relationship


between y and x.


However, inmost cases, the true functional relationship is unknown, and


the experimenter chooses an appropriate function to approximate.


Statistical consideration concerning regression includes various methods.


chart presented in Figure 4.2 shows various regression methods.


The flow


Regression methods are


frequently used to analyze data from unplanned experiments, such as might arise from


observation of uncontrolled phenomena, or historical records.


highly useful in designed experiments.


Regression analysis is also


Generally, the analysis of variance in a designed


experiment helps to identify which factors are important, and regression is used to build a

quantitative model relating the important factors to the response.

Simple linear regression


To determine the relationship between a single regressor variable x and a response

variable y, where x is usually assumed to be a continuous











































Linear Regression


\ *.
*. ~. *~ ~


Regression Analysis


Designed


L -l -i -ii i ^ iii


Multiple Regression


:;Pn::j~~:~~i~a:::::c: ':i:i::' jS
I;~ ~ ~ ~~~* **iiii: filj~~~


A4


Figure 4.2


Various Regression Methods


Source: Ref. 54


Non-experimenta


~~ii~~~iiii~~.~:~i~~~~~~~.~ri~








variable controllable by the experimenter, a simple linear model can be suggested.


relationship between y and


=Po


If the true


x is supposed to be a straight line, the model can be described as


+ PX+


where, po and p,


are unknown constants and


e is a random error with mean zero and variance


The expected value ofy for each value ofx is given by


E(y)


=11


To formulate an estimate for E(y), given sample information is used to construct


estimates, 0o and p ,


of the parameters po and p3


1. The method of least squares is used for


estimating the line and its constants.


The method chooses the prediction line that minimizes


the sum of the squared errors of prediction for all sample points.


The least square estimates


are given as


Sxy


-,1x


where


= x


(Sx)2


2(x


= Zxy


(Sx) (Sy)


Sxy


+Px


X) (y




87


Least square method

To describe the least square method, a graphical representation is provided in Figure


As shown in the figure, the quantity (y


between y and


attributed as error.


+ gx is (yy


- y) represents that the portion of the distance


; that can not be accounted for by the independent variable x and is


Therefore, the sum of squared prediction error from the titled model 9


- ^)2.


From the analysis of the model, all the errors can be formulated as


sum of squares


about the mean


sum of squares


due to regression


sum of squares


for error


Another way to view this equation and explain the variability in the y-values is (55)


Lb2


total variability


in y-values


variability


explained by model


unexplained


variability


To predict y based on the independent variable x, the larger the explained variability

is relative to the unexplained variability, the better the model fits the data, and should lead to


a more precise prediction ofy based on


Correlation


One measure of the strength of the relationship between two variables x and y is called

the coefficient of linear correlation and can be computed as (55)


gn,


_y32