Implementation Process: Appendix G

Choosing the Roadway-Related Strategy

The implementation guides for each emphasis area have provided details of proven, tried, and experimental strategies aimed at reducing crashes, injuries, and deaths resulting from a variety of crash types. For roadway-oriented strategies, a "location" (i.e., system, corridor, section, spot) is usually identified for potential ROR treatment. Once that is done, the question arises as to which of the strategies, or combinations of strategies, is most appropriate. The choice of strategy to be implemented will depend upon many factors – the existing character of the roadway, intersection, and roadside at the site, available right-of-way, strategies already in place, available funding, neighborhood considerations, and jurisdiction treatment and maintenance policies, among others. Thus, the choice of appropriate strategy or strategies cannot be provided by a "formula," but must be based upon safety considerations, engineering analysis, and knowledge.

The usual approach to selecting a strategy, when working in a responsive mode, is to begin by considering the nature of the crash history. For example, if there is an overrepresentation of skidding crashes on wet pavement, one might immediately consider Strategy 15.1 A7 (in the guide addressing run-off-road crashes)—"Provide skid-resistant pavement surfaces" – as a candidate.

However, identifying candidate strategies to study is not always so straightforward. For example, one might find that examination of the "Most Harmful Event" or "Sequence of Events" indicates an overrepresentation of rollover run-off-road crashes at a site. This knowledge might point one to Strategy 15.1 B1 – "Design safer slopes and ditches to prevent rollovers." Even though this strategy would probably be very effective, one may also be able to prevent the vehicle from leaving its lane in the first place, so that the rollover (or fixed object impact) would never occur. In fact, jurisdictional treatment policy may direct that the first priority be given to keeping vehicles on the road. Thus, a range of crash characteristics must be used. Furthermore, all potentially appropriate strategies should be considered, whether they directly or indirectly deal with the problem. The result will often be a list of candidate strategies, which should be studied further for possible use, either singly or in combination. The best method for determining the appropriate combination of strategies to apply is some form of cost-effectiveness or cost-benefit analysis.

This process involves, among other activities, the examination of crash data for the "location" being analyzed; searching for crash factors that might be frequently present or over-represented in the targeted crashes, and choosing strategies which are related to these specific crash factors.

Many attempts have been made over the years to provide tables which relate crash characteristics to possible improvement strategies. Three examples applicable to the implementation guides for the AASHTO Strategic Highway Safety Plan have been provided in:

Appendix G-1: for Run-off-Road Crashes

Appendix G-2: for Crashes involving Trees in Hazardous Locations

Appendix G-3: for Crashes Occurring at Unsignalized Intersections

Appendix G-4: for Crashes Involving Pedestrians

Appendix G-5: for Crashes Occurring at Signalized Intersections

Since the content of state reporting forms is currently quite variable, a standard has been used to identify the factors included in the table. That standard is the Model Minimum Uniform Crash Criteria (MMUCC)¹. Where that standard is not sufficiently comprehensive, reference is also made to the ANSI D20 standard for crash report elements.²

This examination of crash characteristics will often result in the identification of more than one strategy that might be appropriate for a given "location." In these cases, the sum of strategies associated with each of the factors should be considered. For example, if the analysis suggests that there is an overrepresentation of sites with both sharp curves and shoulder drop-off associated with run-off-road crashes, then Strategy 15.1 A8, dealing with shoulder drop off, should be included in the list of candidate strategies along with such strategies as 15.1 A3 through 15.1 A5 (which are associated with the horizontal alignment factor).

The linkage between crash factors and candidate strategies that is provided in the example tables is a simplistic approach to the task of identifying appropriate strategies to use. It is not feasible to provide anything with greater depth or detail. The process that the analyst must go through will be different for each case, due to the complexity of the considerations and factors involved. Therefore, In making the final choice of which of the potential strategies to implement, the analyst would further examine all potential strategies in terms of the factors noted earlier -- the existing character of the roadway and roadside at the site, available right-of-way, strategies already in place, available funding, agency policies and practices, stakeholder requirements, etc.

Finally, it should be remembered that the tables contain only roadway-oriented strategies. As discussed above, there are potentially applicable strategies related to enforcement, public information and education, emergency medical services, and system management, which may also be considered to improve the ROR safety situation. For more on this, please refer back to the section entitled "Related Strategies for Creating a Truly Comprehensive Approach."

¹ Final Model Minimum Uniform Crash Criteria (MMUCC), established by NHTSA and FHWA in August of 1998 (http://www.nhtsa.dot.gov/people/perform/trafrecords/pages/mmucc/mmucc_documents.htm)

² ANSI D20 is a comprehensive data dictionary for crash reports, available from the American Association of Motor Vehicle Administrators (AAMVA)