• Highway Safety Improvement Program (HSIP)
• Intersection Safety
  • Crash Facts
  • Human Factors
  • Alternative Intersection Design
  • Pedestrians & Bicyclists
  • Red-Light Running
  • Roundabouts
  • Rural Intersection Safety
  • Signalized Intersections
  • Unsignalized Intersections
  • Resources
• Local & Rural Road Safety
• Pedestrian & Bicycle Safety
• Roadway Departure Safety
• Speed Management Safety
• Additional Safety Programs & Initiatives
  • Proven Safety Countermeasures
  • Capacity Building
  • Crash Data Improvement Program (CDIP)
  • Crash Tested Hardware
  • Facts & Statistics
  • Focused Approach
  • Geometric Design
  • Highway Safety Manual
  • Manual on Uniform Traffic Control Devices (MUTCD)
  • Motorcycles
  • Newsletters
  • Older Road Users
  • Policy & Guidelines
  • Railway-Highway Grade Crossings
  • Road Safety Audits (RSA)
  • Roadway Safety Data Partnership
  • Safe Routes to School
  • SAFETEA–LU
  • Safety Research
  • Section 402 Highway Safety Funds
  • Tools & Technology
  • Toward Zero Deaths
  • Work Zones

Home > Intersections > Resources

Intersection Safety Implementation Plan Process

Step 2: Expand the Current Approach for Achieving the Crash Reduction Goal

Traditionally, States have relied on one approach to address intersection safety problems - concentrating on improving those intersections with the highest concentrations of frequent and severe intersection crashes. However, in order to meet the intersection crash reduction goal established in Step 1, States likely will have to expand their approach.

Three approaches to implementing intersection improvements probably will be needed to achieve the intersection crash reduction goal, particularly if the goal is designed to achieve a measurable statewide reduction in intersection fatalities or fatalities and incapacitating injuries. The approaches are:

1. Traditional.
2. Systematic.
3. Comprehensive.

If the intersection crash reduction goal is expressed in terms of a reduction in statewide fatalities and a substantial number of the fatalities occur on local roads and intersections, it is probable that these approaches need to be considered for application on both State- and locally-owned intersections. As a rule of thumb, based upon experience gathered from States that have developed intersection safety plans, the relative importance of considering improvements on local intersections to achieve a statewide intersection goal can be reflected in the proportion of intersection fatalities occurring at local intersections as indicated in the following ranges:

• Less than 10 percent of statewide intersection fatalities occur at locally-owned intersections – Minimal importance to include locally-owned intersection improvements to achieve a statewide intersection crash reduction goal.
• Between 10 and 20-25 percent of statewide intersection fatalities occur at locally-owned intersections – Beneficial and probably need to include some local intersection improvements to achieve a statewide intersection crash reduction goal.
• Greater than 20-25 percent of statewide intersection fatalities occur at locally-owned intersections – Necessary to incorporate local intersection improvements to achieve a statewide intersection crash reduction goal.

Traditional Approach

Traditionally, States identify high-crash locations using crash data associated with a highway referencing system and, in some cases, traffic volume information. A formula to rank the locations by some combination of frequency, severity, rate, and crash trend is used to establish candidate locations for improvement. For each candidate location, crash diagrams are developed and studied to determine potential countermeasures for reducing future crash occurrence. A benefit-cost (B/C) analysis usually is performed to determine if the proposed improvement(s) is cost-effective. Those candidate locations with the best benefit-cost ratios may be selected for the limited funding available. Due to the relative high cost of many of these improvements, an average State may implement fewer than 100 traditional safety improvements annually.

While this approach is important and needs to continue, it has minimal impact on reducing substantial numbers of future statewide fatalities and incapacitating injuries. If a State's safety goal is measured by a reduction of statewide fatalities, there is little probability that a fatality would occur at the improvement sites during the next few years, even if the improvements had not been made. The probability of a future fatality occurring is a function of a number of independent variables, many of which safety engineers have no control over, including the following:

• Speed.
• Type of crash.
• Point of impact.
• Type and mass of involved vehicle(s).
• Safety belt usage.
• Type of highway.
• Weather and surface conditions.
• Time of day.
• Type of traffic control.
• Crash location – urban or rural.
• Age and health of drivers and occupants.
• Emergency medical service (EMS) capabilities.
• Distance to nearest hospital.

In addition, statistics from States indicate very few intersections have multiple fatal crashes over a 5-year period. A typical distribution of fatal crashes within a State over a 5-year period is shown in Table 3.

If a fatal crash has occurred at an intersection, there is a relatively low probability that another one will occur within the next few years even if nothing is done to the intersection. If the statewide goal is expressed as a measured reduction of statewide fatalities (or fatalities and incapacitating injuries), then a traditional approach limited to a relatively nominal number of intersection improvements (less than approximately 100 annually) will be insufficient by itself to achieve the goal. Additional approaches to supplement the traditional approach are needed to achieve the intersection crash reduction goal.

Systematic Approach

The systematic approach is the opposite of the traditional approach in that it starts with a set of low-cost, effective countermeasures that the State is comfortable deploying and searches the crash data system to identify intersections where the countermeasures can be deployed cost-effectively. This approach is not limited to the highest crash locations. Typically, it focuses on treating the 3-6 percent of the intersections at which 25-45 percent of the statewide targeted intersection crashes exist.

In the systematic approach, intersection crash data are divided into three levels of information: State or local ownership, urban or rural location, and stop-controlled or signalized. As shown in Figure 2, all combinations of these levels (e.g., state rural stop-controlled intersections, local urban signalized intersections) are used as a basis for analyzing the data. The breakdown of intersection ownership is important since State and local government implementation processes are often quite different. The separation of crashes by urban and rural area is necessary since crash severity (i.e., potential for a fatality) is much greater in rural areas for the same type of crash. The type of traffic control will dictate countermeasure treatment. In addition, the severity of similar types of crashes can differ significantly depending on the type of control (e.g., angle crashes at rural stop-controlled intersections are generally much more severe than angle crashes at rural signalized intersections).

Figure 2: Levels of Information for the Systematic Approach Crash Data Analysis

The cost of a fatality and all injury categories should be used in performing B/C analyses to determine the target crash threshold where it is cost-effective to apply a designated low-cost countermeasure or sets of countermeasures. These costs were updated in a February 5, 2008, US Department of Transportation (USDOT) memo, Treatment of the Economic Value of a Statistical Life in Departmental Analyses,⁸ and are summarized in Table 4.

Crash data analyses to determine if a countermeasure is cost-effective and can be considered for systematic deployment take crash types into consideration. Examples of crash type information needed to evaluate the potential deployment of various countermeasures are shown in Table 5.

Table 5: Targeted Crash Types by Traffic Control, Ownership, and Area

Traffic Control	Ownership	Area	Total Crashes	Angle Crashes	Left Turn Crashes	Dark Crashes	Wet Crashes	Pedestrian Crashes	Speeding Crashes
Stop	State	Rural	X	X	X	X	X		X
Stop	State	Urban	X	X	X	X	X	X	X
Stop	Local	Rural	X	X	X	X	X		X
Stop	Local	Urban	X	X	X	X	X	X	X
Signal	State	Rural	X	X	X	X	X		X
Signal	State	Urban	X	X	X	X	X	X	X
Signal	Local	Rural	X	X	X	X	X		X
Signal	Local	Urban	X	X	X	X	X	X	X

For each of the crash types in Table 5, two key pieces of crash data are needed to perform the analyses:

• The severity of crashes, usually expressed in fatalities per 100 crashes for all of the statewide crashes over the past 5 years. The number of incapacitating injuries per 100 crashes may also be used to measure the impact of a countermeasure on incapacitating injuries.
• The distribution of crashes per intersection using 5 or more years of crash data for all intersections that had at least one crash. For example, this 5-year distribution may show that 25-45 percent of statewide crashes at State, rural, stop-controlled intersections occur in 3-6 percent of the intersections.

An example of typical rates for fatalities per 100 crashes is provided in Table 6. This table shows that rural stop-controlled intersections have the highest severity rates, and that these rates generally increase at night. Pedestrian crashes have a much higher fatality rate than other types of crashes. Crashes at local intersections have a severity similar to, but may be slightly less than, those occurring at State intersections. It is important that each State compute its own values for these severity rates using the most current 5 years of crash data.

Typical distributions for total crashes at State stop-controlled and signalized intersections are provided in Tables 7 and 8.

Two key observations can be made from Tables 7 and 8. In Table 7, if the 513 intersections that had 10 or more crashes were treated with low-cost countermeasures, almost 30 percent of the crashes that occur at State, rural, stop-controlled intersections could be impacted by the countermeasures. In Table 8, if the 31 intersections that had 30 or more crashes were treated with low-cost countermeasures, then over 34 percent of the crashes that occur at State, rural, signalized intersections could be impacted by the countermeasures. Conceptually, this is the essence of the systematic approach – identifying a relatively small set of intersections that comprise a substantial portion of the statewide crash problem, and treat the set with effective, low-cost countermeasures.

When performing the above analyses, it is important that a minimum of 5 full years of crash data be utilized. More years may be used if the data is available in the crash data system and factors that can change exposure (e.g., significant land use changes, traffic volume changes) have not occurred over the crash data period. Three years of data, while acceptable for identifying high-crash locations, is considered too unstable for identifying intersections with lower repetitive crash histories to be considered for systematic deployment of low-cost countermeasures. In addition, each State should define its own threshold levels based upon the data analyses, a State's ability to implement countermeasures, and the intersection crash reduction goal.

Comprehensive Approach

Since poor driving behavior contributes substantially to intersection crashes, it is important to consider initiatives which can improve safe driving through intersections. The comprehensive approach combines low-cost engineering countermeasures with targeted education and enforcement countermeasures. It is not economical to apply the education and enforcement components to a single intersection. The comprehensive approach works best on a corridor or within a specific area (usually defined by municipality boundaries) with a significant number of severe intersection crashes. The most predominant driving characteristics are speeding on approaches to intersections (both stop-controlled and signalized) and red-light running at signalized intersections. To a lesser extent, running Stop signs and pedestrian movement violations may be specific concerns for a given corridor or area.

In all cases where education and enforcement initiatives are to be considered, appropriate low-cost engineering countermeasures should supplement the initiative and be in place before the education and enforcement initiatives begin. Examples of supplemental low-cost countermeasures include appropriate speed limit sign adjustments, traffic calming measures, and traffic signal enhancements (e.g., combined yellow plus all red clearance interval timing adjustments, increasing the visibility of the signal heads).

The State crash data system may be used to identify priority corridors and municipalities with high numbers of intersection crashes. Those 5 to 10 mile sections of highway with the highest number of intersection fatalities and incapacitating injuries over a 5 year period would be candidates for corridor intersection safety improvements. Those municipalities with the highest number of 5-year intersection fatalities and incapacitating injuries (either total, on a per capita basis, or on a VMT basis) can be considered for the area-wide approach.

⁸ http://ostpxweb.dot.gov/policy/reports/080205.htm