U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
|< Previous||Table of Contents||Next >|
The systematic approach is the reverse of the conventional highway safety improvement program approach. The conventional approach starts with the identification of high-crash intersections and then selects countermeasures to impact crash patterns at the intersections. The systematic approach starts with defining a set of specific low-cost countermeasures and searches the crash data base to identify intersections where they can be deployed cost effectively on rural State, stop-controlled intersections.
Crash frequencies per intersection, expected crash severity (fatalities per 100 crashes), type of traffic control at the intersection (STOP sign, traffic signal), countermeasure costs, and the expected crash reductions resulting from the use of low-cost countermeasures all have to be considered to determine the optimum use of limited funds, particularly if the goal is to maximize the reduction of fatal or severe crashes at intersections.
Figure 7 identifies the eight intersection types in which low-cost countermeasures should be considered for application.
Each of these eight categories has a distinct distribution of crashes per intersection and severity (fatalities per 100 crashes) for each type of crash that a given low-cost countermeasure is designed to impact.
A minimum of 5 years of crash data is recommended to identify intersections with higher levels of crashes. Crashes are not uniformly distributed throughout the universe of intersections. A substantial portion of statewide intersection crashes may occur at only a very small number of intersections that have a higher frequency of crashes per intersection. (As an example, 25% of the statewide crash problem at rural stop-controlled intersections may be concentrated on only 3% of the rural stop-controlled intersections which have high frequencies of crashes). Low-cost countermeasures can be targeted to this small number of intersections and impact a substantial portion of the statewide crash problem.
Figure 7: Intersection Categories for Low-Cost Countermeasure Consideration
A typical statewide distribution of rural State, stopcontrolled intersections with one or more crashes over the past 5 years is shown in Table 7. Most States should be capable of developing similar tables from their Crash Data System using basic software such as ACCESS.
|Number of Crashes per Intersection||Number of Intersections||Cumulative||Cumulative|
|100 and greater||-||-||0.00%||-||0.00%|
|50 - 99||4||4||0.08%||266||2.23%|
|30 - 49||5||9||0.19%||448||3.75%|
|20 - 29||29||38||0.78%||1,095||9.16%|
|10 - 19||125||163||3.36%||2,685||22.47%|
|5 - 9||416||579||11.94%||5,372||44.96%|
In the example above, there were 4,849 stop-controlled intersections that had at least one crash in the past 5 years. There are many more intersections on the State system, but the number of intersections with no crashes can not be identified from the crash data base. Overall there were 11,948 crashes at the 4,849 intersections. Of these intersections, 579 (12 percent of total) intersections had five or more crashes per intersection, resulting in 5,372 crashes, or almost 45 percent of all statewide crashes. At a higher threshold, 163 intersections had 10 or more crashes per intersection resulting in 2,685 crashes or over 22 percent of the statewide crashes. Once a minimum crash threshold is established, applying lowcost countermeasures at those intersections with crash frequencies at or above the threshold would impact a substantial portion of all statewide rural crashes at stopcontrolled intersections on State roads.
In addition to identifying intersections with higher frequencies of crashes per intersection, the severity of crashes (fatalities per 100 crashes) can differ significantly by area (i.e., rural or urban) and type of traffic control and needs to be considered. For example, crashes at rural stopcontrolled intersections are typically much more severe than crashes at either urban stop-controlled intersections or any type of signalized intersections.
The definition of severity used in the analysis should be aligned with the intersection goal of the Strategic Highway Safety Plan. If the goal of the Plan is to reduce intersection fatalities the severity should be defined in terms of fatalities per 100 crashes; if the goal is to reduce fatalities and incapacitating injuries, the severity should be defined in terms of fatalities and incapacitating injuries per 100 crashes. The severity of any given crash is dependent on a number of factors, many of which are independent of the intersection characteristics. Factors such as the types of vehicles in the crash, the kinematics of the crash, and the age and gender of the drivers can greatly influence the potential for a crash to result in either property damage or a fatality. Statewide fatalities or fatalities and incapacitating injuries per 100 similar crashes for a given type of intersection and crash type provide a more stable estimate of the severity.
The severity of crashes varies significantly by ownership, urban/rural area, and type of traffic control. Table 8 provides an example of differences in severity (fatalities per 100 crashes) by type of traffic control, area, and ownership for a typical State.
|Traffic Control||Road Ownership||Fatalities per 100 Crashes|
As an example, if the application of low-cost countermeasures at State rural stop-controlled intersections is expected to prevent 100 crashes from occurring, 4.4 lives are projected to be saved. Similarly, if the same low-cost countermeasure applied at State urban stop-controlled intersections is also expected to prevent 100 crashes from occurring, only 1.14 lives are projected to be saved. If a State has a goal to reduce intersection fatalities and has 100 urban and 100 rural stop-controlled intersections each with five crashes per intersection, and can only improve 100 intersections it is clear that improving the rural intersections has more life saving potential; however, if the urban intersections had 20 crashes per intersection the decision to maximize lives saved becomes much more difficult.
|< Previous||Table of Contents||Next >|