Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 55


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 54
54 SECTION VI Special Road User Populations Planning Programs Related "known-effectiveness procedures" are provided in Section IV To Reducing Crashes Involving on "Roadway Segment Programs." Older Drivers, Younger Drivers, Thus, the basic steps in Procedure 3 presented below will be Pedestrians And Bicyclists appropriate for all four of the road user populations covered in this section. The data (e.g., variable values used to define older This section of the guide provides the details of choosing driver crashes and crash types for older drivers vs. pedestrians) treatment strategies for older drivers, younger drivers, pedes- will differ, but the basic procedure will remain the same. The trians or bicyclists, and targeting those treatments to sub- analyst is strongly urged to carefully review the material in each groups of these populations or to locations where their crashes of the pertinent guides before beginning this planning process. occur. As indicated earlier, it is assumed at this point that the These user-population-oriented guides are found within analyst has chosen his/her other emphasis area or areas (e.g., NCHRP Report 500: Guidance for Implementation of the older drivers and/or pedestrians) and has established a stretch AASHTO Strategic Highway Safety Plan. The specific volumes goal. In implementing driver-oriented programs, the estima- pertinent to this section are: tion of program costs is often challenging. In addition to the direct cost of the program, one-time start-up costs and Volume 9: A Guide for Reducing Collisions Involving Older indirect/administrative costs may be substantial, but are not Drivers (9) always addressed in the cost estimation process. Planning of Volume 10: A Guide for Reducing Collisions Involving Pedes- pedestrian and bicycle safety programs is often challenging trians (10) because of limited crash data. In planning pedestrian safety Volume 18: A Guide for Reducing Collisions Involving Bi- improvements, opportunities to improve accessibility under cycles (2008) the requirements of the Americans with Disabilities Act Volume 19: A Guide for Reducing Collisions Involving Young (ADA) should be addressed. Drivers (2007) Four procedures for choosing and targeting treatment strategies were described in the Stage 3 text in Section III. A link to these downloadable guides can be found at http:// Three of those procedures require that the effectiveness (CRF safety.transportation.org/guides.aspx. or AMF) of at least part of the potential treatment strategies be known. However, almost none of the strategies in the Procedure 3 Choosing Roadway guides related to these special road-user populations have User Treatments and Target Subgroups known effectiveness. For that reason, only the details of When Treatment Effectiveness in Procedure 3 will be covered in this section. If AMFs are de- Terms of Crash/Injury Reduction veloped for treatments for these populations, or if the analyst Is Not Known is only interested in examining the few treatments with known AMFs, then the economic-based Procedures 1 or 2 Again, the assumption here is that there is no known level can be used. If AMFs exist for some of the treatments of of effectiveness for the treatment strategies of interest no de- potential interest but not for all (which will likely be the fined CRFs or AMFs. Thus, economic analyses like those that case in the near future), Procedure 4 can be used. While the are the basis for Procedures 1, 2A and 2B, and 4 are not possi- crash types will differ, details of the use of all three of these ble for these treatments. This Procedure 3 is aimed at helping

OCR for page 54
55 the analyst make an educated choice of which treatments will While certainly not always the case, the variables (or similar be most effective in their jurisdiction, and to help the analyst variables) listed in Exhibit VI-1 will be used in this identifica- develop a targeting strategy for the treatment in cases where it tion of target-population crashes. is not to be applied jurisdiction-wide (e.g., where specific user Defining crashes that will guide the choice of treatment strat- subpopulations or roadway locations are to be targeted). In egy and the targeting of these strategies will require crash data general, within each user group, the choice between alterna- that include specific variables and codes on such items as tive treatments will be based on the specific nature of the pop- location of crash (intersection vs. non-intersection), condi- ulation's crash problem, and the choice of target subgroups tion of driver or pedestrian, driver/bicyclist/pedestrian action will be based on the determination of where the crash/injury prior to crash, light condition, etc. Again, the names of problem of interest is found. A discussion of this more general variables and the specific codes needed to conduct these analy- procedure was included above, and the reader should review ses will vary from jurisdiction to jurisdiction. While not all that section. crash types for all treatment strategies related to all four populations are included here, Exhibit VI-2 provides some guidance concerning where example variables related to some Data Needs treatment strategies might be found. Note that in MMUCC- The only required data for Procedure 3 are crash data that compliant databases, the term "non-motorist" will be used for will allow the analyst to (1) isolate crashes involving the spe- both pedestrians and bicyclists. cific user population of interest (e.g., older drivers) and (2) define crash types for this user population which would Procedure suggest strategies and target subgroups. DMV records, and particularly DMV driver history files, may also be useful in As described in Section III, Procedure 3 has two basic steps. planning driver-oriented programs. First, choose the best treatments for the user population of To isolate crashes involving the population of interest, the interest (e.g., the older-driver treatments most likely to be analyst will need to examine the data formats/coding in his/her applicable in a given jurisdiction) from among the set of all crash file to identify variables that can be used in determining treatments presented in the applicable NCHRP Report 500 whether or not a given crash is a "target-population crash." guides. Second, choose the subgroups of users or highway Crash databases often categorize data for a given crash into up locations to which the selected treatments should be applied. to three subfiles (1) general accident/crash variables ("crash"), As described earlier in more detail, the choice of the "best treat- (2) variables for each vehicle in the crash ("vehicle"), and (3) ments" from the listing of many potential user-population variables for each occupant/person in the crash ("person" or treatments can be based on the following factors: "occupant"). The variables needed to determine whether a crash is a "target-population crash" are usually found in the oc- a) The potential treatment judged to be the most effective, cupant/person subfile, but could also be found in the general even given that effectiveness is unknown crash subfile (e.g., a "flag" for all pedestrian crashes) or "vehicle" b) The relative magnitude of the crash types and severity levels subfile (e.g., driver information included with each vehicle that the treatment will affect record). Pedestrians or bicyclists are sometimes classed as a c) The cost of the potential treatments (either jurisdiction- "vehicle type" in the vehicle file, and sometimes as a "person wide, per-mile or per-location) type" in the occupant/person file. (If the jurisdiction's data are d) Other technical or policy considerations compliant with the Model Minimum Uniform Crash Criteria MMUCC these data variables will be in the "Person" subfile.) These factors must be combined in some fashion to decide In short, crash files differ from jurisdiction to jurisdiction. which treatment to choose. While there are multiple ways of Population Type Variable Crash Database Subfile Older Drivers or Younger Person Type Person/Occupant Drivers Driver Age Person/Occupant or Vehicle Driver Date of Birth Person/Occupant or Vehicle Pedestrian or Bicyclist Person Type Person/Occupant Vehicle Type Vehicle Crash Type (First Harmful Crash Event) Exhibit VI-1. Crash variables and subfile location by population type.

OCR for page 54
56 Crash Type/Issue Variable Crash Database Subfile Intersection vs. Non-intersection Relation to Junction Crash Type of Intersection Crash Traffic Control Device Type Crash or Vehicle Nighttime/Reduced Visibility Light Condition Crash Weather Crash Roadway Lighting Crash Lane Departure (Potentially Related Accident/Crash Type Crash to Pavement Markings) Manner of Collision Crash Sequence of Events Vehicle First Harmful Event Crash Most Harmful Event Vehicle Crash Location (Off-road) Crash Crashes Associated with Medical Driver Condition Person Conditions Occupant Restraint Use Occupant Protection System Person Use Work Zone Work Zone Related Crash Roadway Condition Crash Pedestrian "Walking along Pedestrian (or Non-Motorist) Person Roadway" Crashes Action Prior to Crash Speed-related Crashes Driver Action Prior to Crash Person (or Vehicle) Violation Indicated Person (or Vehicle) Contributing Circumstances Person (or Vehicle) Crash Location (for Targeting County Crash Treatments) City Crash Route Crash Milepost Crash Longitude/Latitude Crash Block Address Crash Speed Limit (for Use in Developing Speed Limit Crash Cost per Crash) Exhibit VI-2. Crash variables and subfile location by crash type/issue. making this choice, the following represents one such pro- population. Information on economic cost per crash cedure. severity level can be found in Crash Cost Estimates by Maximum Police-Reported Injury Severity Within Selected 1. Prioritize the specific user-population problem(s) to be Crash Geometries (22). Here, instead of using severity cost addressed. by crash type as is done in roadway-program analyses This is related to Factor b in the above list. Here, the covered in earlier sections, the analyst will use the basic initial issue is whether to treat older driver, younger crash cost by police-reported severity level (i.e., driver, pedestrian or bicyclist crashes. This prioritization K,A,B,C,O). Exhibit VI-3 below presents those costs per will be based on the frequency and severity of the specific crash. Costs for combinations of crash severity levels (e.g., types of user-population crashes occurring in an analyst's K+A crashes) are presented in that report (22). This jurisdiction. Crashes specific to a given user population were defined in the table above. For each user population, Crash Severity Speed Limit Comprehensive the analyst could begin the process by analyzing 3 to 5 Category Cost/Crash* years of crash data to determine the frequency of each < 45 mph $3,622,200 Fatal (K) population. However, since some crashes for some pop- > 50 mph $4,107,600 < 45 mph $195,700 ulations are more severe than others, total crash Serious Injury (A) > 50 mph $222,300 frequency alone does not provide the complete answer. Moderate injury (B) < 45 mph $62,200 While an alternative is to restrict the analysis to only fatal > 50 mph $91,600 < 45 mph $40,100 and serious-injury crashes, this will severely limit the Minor Injury (C) > 50 mph $49,500 crash sample, and will also omit a large component of the No Injury (O) < 45 mph $7,000 > 50 mph $7,800 crash problem non-serious injury and no-injury *Crash cost in 2001 dollars crashes. A better solution is to weight each crash for a given user population by an economic cost based on its Exhibit VI-3. Crash cost by crash severity and posted severity, and then accumulate the total crash cost for each speed limit (22).

OCR for page 54
57 analysis of total crash cost will provide the analyst with 3. Rate the possible treatments based on estimated effec- overall information on which user population is most im- tiveness. portant in his/her jurisdiction. Since this procedure deals with treatment strategies For the chosen user population, the analyst could then with unknown effectiveness, this appears to be impossible. conduct additional analyses of "critical crash types" for However, for a given set of possible treatments for a par- that population by producing crash-type distributions and ticular user group, it may be possible to make a judgment weighting each crash type by the cost per crash. This could concerning which treatment strategy would be expected to be done either by using the costs for the 22 crash types be most effective. For example, strategies related to chang- presented in the above report, or by developing severity ing the roadway may be more effective, in general, than distributions within each crash type and weighting the strategies related to education (but, of course, will affect individual severity-level frequencies by the cost estimates only those users at the treated locations). At times, this will above. This analysis will then produce a listing of potentially clearly be a very difficult judgment to make. treatable crash types for the chosen user population that can 4. Choose "best" treatment(s) by considering estimated be sorted by crash frequency or total crash cost, thus pro- effectiveness, cost, and other technical and policy con- viding a ranked listing. For the higher-ranked crash types, siderations. the analyst can then conduct additional analyses to deter- The analyst will then combine the output of the steps mine more of the specifics of the crash circumstances (e.g., above with at least two other factors in making a final nighttime vs. daytime distributions of total crash cost). decision on which treatment(s) to implement the cost These additional "drill-down" analyses should be designed of the treatment and other technical and policy consid- to provide additional information that could lead to the erations. Unfortunately, there are no good guidelines choice of one treatment over another (e.g., intersection for how to "weight" the different factors. While problem lighting will affect nighttime older-driver crashes at inter- size (total crash cost) and assumed treatment effective- sections, and traffic calming measures on road sections are ness are key factors, there may be technical, policy, and more likely to affect locations with higher speeds, as defined cost considerations that will remove certain treatments by either speed limit or speeding as a contributing factor). from consideration even if they are felt to be effective. 2. Identify possible treatments for use for each high-priority The analyst will have to choose the final treatments crash type. based on best judgment. The procedure outlined above The analyst will then review the pertinent NCHRP Re- will at least ensure that the major factors in the decision port 500 guides and list treatments that would be most ap- are clearly defined. The output of this step will be one or propriate for each of the high-priority crash types identi- more chosen treatments, with the nature of the treat- fied in the above step. The choice should be limited to ment defining the specific crash types more likely to be those treatment strategies that are classified as proven or affected. tried in the guides. If not already conducted in the "drill- 5. Target the chosen treatments to the user populations down" analysis in the preceding step, more specific infor- where the problem is found. mation on the total crash cost related to each potential In some cases, treatment strategies related to user pop- treatment strategy could be developed by specifying the ulations will be implemented jurisdiction-wide. In other crash types that are most likely to be affected by each cases, it may be desirable to target the treatment to either strategy (e.g., pedestrian-crossing crashes at higher-speed a subgroup of the user population or to specific locations intersections as targets for intersection traffic calming (e.g., specific counties, route sections, or intersections). If treatments), producing crash frequencies for each speci- a given strategy can be linked to a specific crash type or fied crash type, and multiplying the frequencies by cost types, choosing high-priority subgroups for targeting can per crash. For some strategies, the NCHRP Report 500 se- be done using similar procedures noted above for choos- ries presents information concerning which crash types ing treatments. Here, the user-population crashes within are most likely affected by that treatment strategy. How- each crash type would be divided among all potential user ever, for other user-population strategies, it will not be subgroups (e.g., pedestrian crashes would be divided into possible to define one or more specific crash types for a age groups), and crash frequency or total crash cost would given potential strategy (e.g., education programs for be calculated for each subgroup, producing a ranking drivers and pedestrians, resource centers to promote safe based on problem size. If treatment-cost estimates can be mobility choices for older drivers). In these cases, the an- made for each subgroup, the total crash cost and treat- alyst will have to make some judgment concerning the ment cost can be combined to provide an indication of relative size of the crash problem that could potentially be which subgroup might produce the largest payoff per affected by these strategies. treatment dollar spent.