1. CE 4640: Transportation Design
Prof. Tapan Datta, Ph.D., P.E.
Fall 2002

2. Highway Safety: Facts & Figures
National crash statistics for 2000*
41,821 people were killed in 6,394,000 reported motor vehicle traffic crashes
3,189,000 people were injured
4,286,000 crashes were PDO type
*Source: National Highway Traffic Safety Administration (NHTSA)

3. Highway Safety: Facts & Figures
Source: National Highway Traffic Safety Administration (NHTSA)

4. Highway Safety: Facts & Figures
Source: National Highway Traffic Safety Administration (NHTSA)

5. Highway Safety: Facts & Figures
Source: National Highway Traffic Safety Administration (NHTSA)

6. Highway Safety: Facts & Figures
Michigan crash statistics for 2000*
1,382 people were killed in 424,852 reported motor vehicle traffic crashes
121,826 people were injured
336,572 crashes were PDO type
*Source: Michigan Office of Highway Safety Planning (MOHSP)

7. Highway Safety: Facts & Figures
Michigan crash facts for 2000*
Every 1.2 minutes a traffic crash occurs
One person is killed in a crash every 6.3 hrs
One person is killed in an alcohol-related crash every 19.1 hrs
One driver under age 21 is in a fatal crash every 28.2 hrs
One person is injured in a crash every 4.3 minutes
One pedestrian is injured every 3.6 hrs
One bicyclist is injured every 4.7 hrs
*Source: Michigan Office of Highway Safety Planning (MOHSP)

8. Goals of a Traffic Engineer
Can be one or more of the following:
Reduce the frequency or rate of traffic crashes
Reduce the frequency or rate of injury crashes
Reduce the frequency or rate of fatal crashes
Reduce the frequency or rate of specific crash categories, such as, alcohol-related, speeding, older-driver crashes

9. Elements of Highway Safety
Driver
Vehicle
Roadway
Traffic Engineer
has no control

10. Safety Improvement Approaches
Reducing crash occurrence
Installing safety measures
Correcting hazardous roadway features
Improving driver skills by training
Reducing the severity of crashes
Proper geometric design, guardrail, median barrier, breakaway sign post
Improving crash survivability
Vehicle safety features, like air bag, seat belts, energy-absorbing bumper

11. Safety Improvement Approaches
Programmatic safety efforts
Federal and State programs to promote safety on a policy level
State vehicle inspection program
National speed limit
National 21-year-old drinking age
Federal vehicle design standards
Design aspects of safety
Safer design of roadways
Horizontal and vertical alignment
Roadside design
Median barriers
Gore areas

12. Highway Safety Improvement Program (HSIP)
PLANNING COMPONENT
IMPLEMENTATION COMPONENT
EVALUATION COMPONENT

13. Overview of the Highway Safety Improvement Program
PLANS FOR THE TOTAL HIGHWAY SYSTEM
PLANNING
AND
DESIGN
CONSTRUCTION
SAFETY
OPERATION AND MAINTENANCE
HIGHWAY SAFETY IMPROVEMENT PROGRAM (HSIP)
ADMINISTRATIVE DECISIONS
DESIGN STANDARDS, ETC.
CONCERNING GOALS, OBJECTIVES
PLANNING COMPONENT
IMPLEMENTATIONCOMPONENT
EVALUATION COMPONENT
Overview of the Highway Safety Improvement Program

14. IMPLEMENTATION COMPONENT
PLANNING COMPONENT
PROCESS 1
COLLECT AND MAINTAIN DATA
PROCESS 2
IDENTIFY HAZARDOUS
LOCATIONS AND ELEMENTS
PROCESS 3
CONDUCT ENGINEERING STUDIES
PROCESS 4
ESTABLISH PROJECT PRIORITIES
IMPLEMENTATION COMPONENT
PROCESS 1
SCHEDULE AND IMPLEMENT
SAFETY IMPROVEMENT PROJECTS
EVALUATION COMPONENT
PROCESS 2
DETERMINE THE EFFECT OF
HIGHWAY SAFETY IMPROVEMENTS
Highway Safety Improvement Program at the Process Level

15. DEFINE THE HIGHWAY LOCATION REFERENCE SYSTEM
PROCESS 1. COLLECT AND MAINTAIN DATA
SUBPROCESS 1
DEFINE THE HIGHWAY LOCATION REFERENCE SYSTEM
SUBPROCESS 2
COLLECT AND MAINTAIN
CRASH DATA.
SUBPROCESS 3
COLLECT AND MAINTAIN
TRAFFIC DATA.
SUBPROCESS 4
COLLECT AND MAINTAIN
HIGHWAY DATA.
PLANNING COMPONENT
PROCESS 2. IDENTIFY HAZARDOUS LOCATIONS AND ELEMENTS
PROCESS 3. CONDUCT ENGINEERING STUDIES
SUBPROCESS 1
COLLECT AND
ANALYZE DATA.
SUBPROCESS 2
DEVELOP CANDIDATE
COUNTERMEASURES
SUBPROCESS 3
DEVELOP PROJECTS
PROCESS 4. ESTABLISH PROJECT PRIORITIES

16. Highway Safety Improvement Program at the Subprocess Level
PROCESS 1. SCHEDULE AND IMPLEMENT SAFETY IMPROVEMENT PROJECTS
IMPLEMENTATION
COMPONENT
SUBPROCESS 1
SCHEDULE PROJECTS
SUBPROCESS 2
DESIGN AND
CONSTRUCT PROJECT
SUBPROCESS 3
CONDUCT OPERATIONAL
REVIEW
PROCESS 1. DETERMINE THE EFFECT OF HIGHWAY SAFETY IMPROVEMENTS
SUBPROCESS 1
PERFORM NON-CRASH BASED PROJECT EVALUATION
SUBPROCESS 2
PERFORM CRASH BASED PROJECT EVALUATION
EVALUATION
COMPONENT
SUBPROCESS 3
PERFORM PRGRAM
EVALUATION
SUBPROCESS 4
PERFORM ADMINISTRATIVE
EVALUATION
Highway Safety Improvement Program at the Subprocess Level

17. Crash Data Collection
Police Officer collects necessary crash information from the site
Traffic Crash Report Form (UD-10) is used to gather:
Date, time, location information
Weather, pavement, lighting condition
Driver information
Injury, fatality information
Type and severity of crash
A hand sketch showing the positions of vehicles at the time of crash
Other information

18. Traffic Crash Report Form (UD-10)

19. Crash Data Recording
Data collected in UD-10 Forms are entered into the computer system to develop a statewide crash database
Local government departments may develop their own data system
For example, SEMCOG’s Comprehensive Analysis Safety Tool (CAST)

20. A Sample CAST Output

21. A Sample Collision Diagram

22. Typical Traffic Crash Location File and Crash Location Index Card

23. Computerized Spot Map

24. Process 2: Identify Hazardous Locations and Elements
Purpose
To identify hazardous spots, sections, and elements based on the crash, traffic and highway data obtained from Process 1

25. Procedures Involved in Identification
Procedure 1- Frequency Method
Procedure 2- crash Rate Method
Procedure 3- Frequency Rate Method
Procedure 4- Rate Quality Control Method
Procedure 5- Crash Severity Method
Procedure 6- Hazard Index Method
Procedure 7- Hazardous Roadway Features Inventory Method

26. Hazardous highway locations may or may not be high-crash locations
It is important for the highway agency to also consider the identification of locations with a potential for high-crash numbers or severity

27. Time Consideration
Time period should be short enough to identify sudden changes in crash patterns
Time period should be long enough to assure reliability in identifying hazardous locations
Multiples of one year are preferred
Segment Length Considerations
Spot
Section
In both cases should have consistent characteristics of:
Geometrics
Traffic volumes
Condition

28. Data Input

29. Procedure 1 - Frequency Method
Used to identify and rank locations on the basis of number of crashes
This method is the easiest to apply and does not require the use of traffic volume data
Used by many agencies to select the initial group of high crash locations for further analysis
A critical value must be established for location selection (such as 9 or more crashes per year)

30. Advantages
Effective as a tool for providing continuous monitoring of the crash situation in an area
Provides simple, direct method for identifying hazardous locations
Disadvantages
No consideration of exposure
Does not account for crash severity
Does not give consideration to locations with a high potential for crashes, but with no past crash experience

31. Procedure 2 - Crash Rate Method
It combines the crash crash with the various exposure factors
Vehicle Miles of Travel (VMT)
Population
Registered Vehicles

32. Crash Rate
Crash Rates are calculated for a location, segment or an area based on:
Vehicle Miles of Travel (VMT)
Crash Rates are also calculated for an area based on:
Population
Registered Vehicles

33. Crash Rate Calculation
Crash Rate at a location, Rsp=
Freq. of Crashes*106
(365)(T)(V)
Crashes per million vehicles
where T = Period of study (years)
V = Average Annual Daily Traffic (For intersection, sum of all approach volumes)

34. Crash Rate Calculation
Crash Rate at a segment, Rse=
Freq. of Crashes*106
(365)(T)(V)(L)
Crashes per million vehicle miles of travel
where T = Period of study (years)
V = Average Annual Daily Traffic
L = Length of the section (miles)

35. Crash Rate Calculation
Crash Rate for an area =
Freq. of Crashes*106
VMT
Crashes per million Vehicle Miles of Travel
Freq. of Crashes*1000
Population
Crashes per 1,000 Pop
Freq. of Crashes*1000
Registered Vehicles
Crashes per 1,000 Regd. Vehicles

36. Advantages
Combines the use of an exposure factor (traffic volume) and a frequency factor
Remains a relatively simple, direct method
Disadvantages
May over represent hazard at locations with very low traffic volumes
Requires additional data
Does not account for crash severity
Does not give consideration to locations with a high potential for crashes, but with no past crash experience

37. Example Problem

38. Example Problem Using Frequency Method for Total Crashes:
Rank Intersection Total Crash Freq.
1 Plymouth Road & 50
Middlebelt Road
2 Crook Road & 38
Auburn Road
3 Livernois Road & 12

39. Example Problem Crash Rate at a location, Rsp= Freq. of Crashes*106
Using Crash Rate Method for Total Crashes:
Crash Rate at a location, Rsp=
Freq. of Crashes*106
(365)(T)(V)
Crashes per million vehicles
where T = Period of study (years)
V = Average Annual Daily Traffic (For intersection, sum of all approach volumes)

40. Example Problem Total Crash Rate at Crook & Auburn 38*106
(365)(1)(35,700)
Crashes per million vehicles
= 2.92 Crashes per million vehicles
Similarly, we calculate crash rates for other locations.

41. Example Problem Using Crash Rate Method for Total Crashes:
Rank Intersection Total Crashes per Million Vehicles
1 Crook Road &
Auburn Road
2 Plymouth Road & 2.82
Middlebelt Road
3 Livernois Road & 1.29

42. Procedure 3 - Frequency Rate Method
Normally applied by first selecting a large sample of high crash locations based on a “number of crashes”
Then, crash rates are computed and the locations are priority ranked by crash rate
A some what different procedure was developed to compare the dual influence of frequency and rate in a matrix pattern (as shown in the next slide)

43. Multidimensional Crash Data Analysis Matrix
Priority 1
Crash Rate
Priority 2
Crash Frequency

44. Frequency Rate Matrix for Total Crashes for the Example Problem
Priority 1
Priority 2
Total Crashes per Million Vehicle Miles
Priority 3
Crash Frequency

45. Advantages
Alleviates the need to calculate rates at every crash location
Uses both frequencies and rates to assess hazard
Reduces the exaggerated effect of the crash rate on low volume roads and the exaggerated effect of high frequencies at high-volume intersections
Disadvantages
May require considerable funds and manpower
More complex
Does not account for crash severity
Does not give consideration to locations with a high potential for crashes, but with no past crash experience

46. Procedure 4 - Rate Quality Control Method
Utilizes a statistical test
Determines whether the crash rate at a location is significantly higher than a predetermined average rate for locations of similar characteristics
In this method, the crash rate at a location is compared to a “critical rate”

47. Procedure 4 - Rate Quality Control Method
The equation for calculating the critical rate is as follows:
Rc = Ra + K (Ra / M)1/2 + 1/(2M)
where,
Rc = Critical rate for spot or section
Ra = Average crash rate for all spots of similar characteristics or on similar road types
M = Millions of vehicles passing over a spot or millions of vehicles miles of travel on a section
K = A probability factor

48. P (Probability)
K-value
The most commonly used K values are (P = .005) and (P = 0.05)
Advantages
Reduces the exaggerated effect of the crash rate on low volume roads and the exaggerated effect of high frequencies at high-volume urban intersections
Flexible enough to accommodate changing crash patterns
Allows for statistical reliability in identifying location

49. Disadvantages
Relatively complex
Manual application is time consuming and expensive
Does not take severity of crashes into account
Does not give consideration to locations with a high potential for crashes, but with no past crash experience

50. Procedure 5 - crash Severity Method
Description
Used to identify and priority-rank high-crash locations
Some states consider only injury and fatality crashes in identifying
Other states apply weighting factors to crash based on their severity and then compute some form of severity index

51. Crash severity are often classified by NSC within the following five categories
Fatal Crash - one or more deaths
Type-A Injury Crash - Bleeding wound, distorted member
Type-B Injury Crash - Bruises, abrasion, swelling, limping
Type-C Injury Crash - Involving no visible injuries but complaint of pain (probable injury)
PDO Crash - Property Damage Only

52. Equivalent Property Damage Only (EPDO) Method
The equivalency factors vary by state. The formula below is used by Kentucky:
EPDO = 9.5 (F+A) (B+C) + PDO
where, F = No. of fatal crashes
A = No. of A-Type injury crashes
B = No. of B- Type injury crashes
C = No. of C- Type injury crashes
PDO = No. of PDO crashes

53. Advantages
Accounts for the severity of crashes
Highly applicable to rural areas, where high percentages of severe crashes occur
Disadvantages
The severity of a crash is highly dependent on many factors which are unrelated to the highway location (i.e. age and health of passengers, type of vehicle involved, etc.)
Does not consider locations with a high potential for crashes

54. Procedure 6 - Hazard Index Method
It employs a formula to develop a rating index for each suspect site
Factors used in the formula are
No. of crashes per year
Crash rate
Crash severity
Sight distance
Volume/capacity ratio
Traffic conflicts
Erratic maneuvers
Driver Expectancy
Information system deficiencies

55. Hazard Index Method Example

56. Advantages
Comprehensive use of numerous factors related to locational hazards
Highly adaptable, factors which do not apply or are not available may be deleted from analysis
Considers both crash data and variables which indicate a high potential for crashes
Disadvantages
Large amounts of information are necessary
Deletion of too many factors from analysis reduces its effectiveness
Requires considerable expertise in highway safety and human factors
May require data that is not readily available

57. Procedure 7 - Hazardous Roadway Features Inventory
It is one way of selecting sites with potential for high-crash severity or numbers
Based on comparison of existing features with safety and design standards

58. Examples of Hazardous Features
Blunt-end guardrail barrier terminals
Narrow bridges
Steep roadside slopes
Rigid roadside objects
Narrow lanes and shoulders
Slippery pavements
Sharp radii on horizontal curves and ramps
Hazardous highway-railroad grade crossings

59. Advantages
Considers locations that are potentially hazardous (even though large number of crashes may not have been observed
Considers locations (e.g., railroad grade crossings, roadside hazards) which have a potential for high-severity crashes
Disadvantages
Can require large amounts of data
Requires personnel with experience in highway safety
Improvement expenditures must be justified on some basis other than reduction in crash experience

60. Left-turn Head-on Collisions:
Probable Causes
Restricted sight distance
Too short amber phase
Absence of special left-turning phase
Excessive speed on approaches

61. Left-turn Head-on Collisions: Studies to be Performed
Review existing intersection channelization
Volume count for through & left-turn traffic
Review signal phasing
Study the need for special left-turn phase
Perform spot speed study

62. Left-turn Head-on Collisions: Possible Countermeasures
Provide adequate channelization
Install traffic signal if warranted by MUTCD
Increase amber phase
Provide special phase for left-turning traffic
Prohibit left turns
Reduce speed limit on approaches
Provide all-red phase

63. Rear End Collisions at Unsignalized Intersections:
Probable Causes
Improper channelization
High volume of turning vehicles
Slippery surface
Inadequate intersection warning signs
Excessive speed on approaches
Inadequate roadway lighting

64. Rear End Collisions at Unsignalized Intersections:
Studies to be Performed
Review existing channelization
Volume count for through & turning traffic
Check skid resistance
Perform spot speed study
Check for adequate drainage
Check roadway illumination

65. Rear End Collisions at Unsignalized Intersections:
Possible Countermeasures
Create right or left turn lanes
Increase curb radii
Prohibit turns
Provide “Slippery When Wet” sign
Increase skid resistance
Improve drainage
Reduce speed limit
Improve roadway lighting

66. Angle Collisions at Signalized Intersections:
Probable Causes
Restricted sight distance
Inadequate roadway lighting
Poor visibility of signal
Excessive speed on approaches

67. Angle Collisions at Signalized Intersections: Studies to be Performed
Volume count on all approaches
Field observations for sight obstructions
Review signal timing
Check roadway illumination
Perform spot speed study

68. Angle Collisions at Signalized Intersections: Possible Countermeasures
Remove obstructions to sight distance
Increase amber phase
Provide/increase all-red interval
Prohibit curb parking
Install backplates, larger heads for signals
Improve location of signal heads
Reduce speed limit at approaches

69. A Project Example
An improvement project was undertaken by WSU for AAA, Michigan
One of the study stretch was Eastern Corridor
Problems were identified through studies and analyses
Possible countermeasures were suggested

70. Sample Intersection Before Improvement
Eastern Avenue & Franklin Street

71. Problems Identified
A few of the intersections without a left-turn lane had a queue and associated delay
Signals were not properly located for clear visibility
Signal heads were small 8
Too short all-red intervals at a few locations

72. Suggested Countermeasures
Exclusive left-turn lanes were added wherever needed
Replaced existing span wire mounted signal configuration with box span installation
Relocated signal heads to improve visibility
Replaced 8” signal heads with 12” signal heads       

73. Suggested Countermeasures
Implemented longer all-red intervals at locations where needed (Eastern Avenue 2.0 seconds; minor streets 2.0 seconds)
Installed secondary post mounted signal heads to improve visibility for left turn traffic
Installed back plates on signals to improve visibility

74. Sample Intersection After Improvement
Eastern Avenue & Franklin Street (After)

75. Non-Accident Measures
Reduce the number of conflicts at an intersection, by prohibiting turns or movements
Definition of Traffic Conflict:
an evasive action taken by the driver of a vehicle to avoid an impending collision