Download Behavioral, Operational and Safety Effects of Red-Light Cameras at Signalized Intersections in ... PDF

TitleBehavioral, Operational and Safety Effects of Red-Light Cameras at Signalized Intersections in ...
LanguageEnglish
File Size2.6 MB
Total Pages110
Document Text Contents
Page 1

Behavioral, Operational and Safety Effects of Red-Light Cameras at Signalized
Intersections in Alabama


by


Fatemeh Baratian Ghorghi





A dissertation submitted to the Graduate Faculty of
Auburn University

in partial fulfillment of the
requirements for the Degree of

Doctor of Philosophy


Auburn, Alabama
December 10, 2016






Keywords: red light camera, traffic violation, driver behavior,
operation, crash probability, fine structure





Copyright 2016 by Fatemeh Baratian Ghorghi



Approved by


Huaguo Zhou, Chair, Associate Professor of Civil Engineering
Rod E. Turochy, Associate Professor of Civil Engineering

Wesley C. Zech, Professor of Civil Engineering
Jeffrey J. LaMondia, Associate Professor of Civil Engineering

Ana Franco-Watkins, Associate Professor of Psychology

Page 2

ii

Abstract



Statistics reveal that from 2007-2011 an average of 751 people died each year in red-light running

(RLR) crashes in the U.S. Past studies showed that red light cameras (RLCs), as an enforcement

countermeasure, can lower RLR fatalities at signalized intersections. Currently, approximately 430

individual communities run RLC programs in the U.S. and over 40 intersections in Alabama are

equipped with these cameras. As more RLCs are installed at intersections in Alabama,

understanding their effects and how to best implement them is of growing importance. While

extensive research has investigated the safety effects of the system, very little work has been done

to investigate the impacts of RLCs on driver behavior and intersection operation. To date, very

few study has evaluated the effects of RLCs in Alabama. The primary objective of this study is to

fill the research gap by evaluating the effectiveness of RLC program, in terms of safety, operation,

and driver behavior, while also developing a novel fine structure for RLR traffic violations. In the

first step, the complete process of extracting RLR crash data from Critical Analysis and Reporting

Environment is presented to identify target crashes. More importantly, an extensive field

observation is conducted to collect drivers’ responses to clearance intervals at four intersections

with RLCs and four intersections without RLCs. The increase in the intersection delays due to the

presence of RLCs can be estimated. The results indicate a higher tendency to stop and a longer

delay at intersections equipped with RLCs. Furthermore, a comparison among clearance lost time

values, collected in the field and estimated using the Highway Capacity Manual method and

Alabama Department of Transportation’s Traffic Signal Design Guide and Timing Manual

method, demonstrates that both manuals overestimate the intersection's capacity. An adjustment

factor is estimated and recommended for improving accuracy of both methods. In the last step of

the research, a novel method is developed to determine a basis for RLR fines by considering the

cost of a potential RLR crash and its resulting delay, which is the first of its kind reported in the

literature. Various statistical tests and simulation models are used to accomplish the objectives of

this study.

Page 55

43



logic. The result was then merged with the combination of variables defined in steps 3 through 8

by AND logic. In order to confirm that the crashes were caused due to a vehicle running the red

light, those crashes that occurred due to driving under the influence of drugs or alcohol and crashes

related to emergency vehicles or in police pursuit were filtered out. This was done through steps 6

and 7. The CARE “create filter” work screen is shown in Figure 5.2. More information about how

to create a filter is provided in the Care Filter Catalog (2010).


Figure 5.2 Red-Light-Running Crash Filter



To properly establish the relationship between crashes and enforcement program, efforts

were made to assign at-fault vehicles to the given approach. This was done by controlling direction

of travel and the approach street for every crash. The monitored approaches at RLC intersections

are as follows:

Site 1: At the intersection of Gateway Drive and Pepperell Parkway, there are three RLCs facing

northbound lanes of Gateway Drive, eastbound lanes of Pepperell Parkway, and

westbound lanes of Pepperell Parkway.

Page 56

44



Site 2: Traffic on the northbound lanes of Gateway Drive and the eastbound lanes of Frederick

Road are subject to being monitored by two RLCs located at the intersection of Gateway

Drive and Frederick Road.

Site 3: There are two RLCs located at the intersection of Gateway Drive and Interstate Drive,

focused on the northbound and the southbound lanes of Gateway Drive.

Site 4: There is only one RLC at the intersection of Fox Run Parkway and Samford Avenue

facing the westbound lane of Samford Avenue.

Figure 5.3 shows the monitored approaches at four RLC intersections.




Figure 5.3 Positions of the RLCs at Treated Intersections



Site 1 Site 2

Site 3 Site 4

Page 109

97






Table A.5 Simple Method Input Data –RLR Crashes

Intersection
Years
Before

Years
After

Crashes before
K(j)

Crashes After
L(j)

rd(j) rd(j)×K(j) rd(j)2×K(j)

Pepperell Prkwy & Gateway Dr 3 2.5 4 2 0.83 3.33 2.78

Frederick Rd & Gateway Dr 3 2.5 4 0 0.83 3.33 2.78

Interstate Dr & Gateway Dr 3 2.5 5 6 0.83 4.17 3.47

Fox Run Pkwy & W Point Ave 3 2.5 1 1 0.83 0.83 0.69

Sum 14 9 11.67 9.72








Table A.6 Simple Method Input Data –RE Crashes

Intersection
Years
Before

Years
After

Crashes before
K(j)

Crashes After
L(j)

rd(j) rd(j)×K(j) rd(j)2×K(j)

Pepperell Prkwy & Gateway Dr 3 2.5 3 4 0.83 2.50 2.08

Frederick Rd & Gateway Dr 3 2.5 2 2 0.83 1.67 1.39

Interstate Dr & Gateway Dr 3 2.5 1 4 0.83 0.83 0.69

Fox Run Pkwy & W Point Ave 3 2.5 1 0 0.83 0.83 0.69

Sum 7 10 5.83 4.86

Page 110

98






Table A.7 Simple Method Input Data –Injury Crashes
Intersection Years

Before
Years
After

Crashes before
K(j)

Crashes After
L(j)

rd(j) rd(j)×K(j) rd(j)2×K(j)

Pepperell Prkwy & Gateway Dr 3 2.5 1 2 0.83 0.83 0.69

Frederick Rd & Gateway Dr 3 2.5 2 1 0.83 1.67 1.39

Interstate Dr & Gateway Dr 3 2.5 1 4 0.83 0.83 0.69

Fox Run Pkwy & W Point Ave 3 2.5 1 1 0.83 0.83 0.69

Sum 5 8 4.17 3.47









Table A.8 Simple Method Input Data –PDO Crashes
Intersection Years

Before
Years
After

Crashes before
K(j)

Crashes After
L(j)

rd(j) rd(j)×K(j) rd(j)2×K(j)

Pepperell Prkwy & Gateway Dr 3 2.5 6 4 0.83 5.00 4.17

Frederick Rd & Gateway Dr 3 2.5 4 1 0.83 3.33 2.78

Interstate Dr & Gateway Dr 3 2.5 5 6 0.83 4.17 3.47

Fox Run Pkwy & W Point Ave 3 2.5 1 0 0.83 0.83 0.69

Sum 16 11 13.33 11.11

Similer Documents