Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
32 Conclusions This research aimed to develop a set of recommendations for evaluating the safety implications of selected design element trade-offs. The research team used an expert-panel approach where prior research was reviewed and discussed along with the models developed herein. The team discussed and com- pared past work with that completed here and recommended a set of AMFs that could be used in determining the safety effects from the change in the values of a design element. Their final recommendations were for shoulder width and median width for four-lane roads with 12-ft lanes. The available data did not allow for the development of additional recom- mendations even though the presence of median barrier was also considered. The values recommended here are higher than those proposed in the HSM mainly because they address all crashes rather than only crashes related to the specific element. This fact explains the larger magnitude of these AMFs since they capture the effect of a larger number of crashes. Two sets of recommended AMFs for shoulder width were developed that could be used based on whether the roadway is divided. Each set addresses the effect of the shoulder width on the potential crash occurrence for the total number of crashes and represents the relative change from using the specific value. Through the expert panel approach, the research team concluded that these AMFs were appropriate and rea- sonable to use for estimating the effect of the shoulder width on crash occurrence. Increasing the shoulder width by 1 ft for undivided highways effects an approximately 6% crash reduction, while for divided highways the reduction is 5%. These values are in accordance with past work and demonstrate the positive effect of shoulder width on crash occurrence. A single set of AMFs is recommended for the median width, for multi-vehicle crashes for divided roadways, since this vari- able was only present in the model for multi-vehicle crashes. The research team through its expert-panel approach deter- mined that this factor was reasonable and recommended its use. The effect of median width on crashes is approximately an 8% reduction with every 10-ft increase in median width. An AMF for all crashes could be developed by assuming that the AMF for single-vehicles is 1.00 and estimating a weighted average using the percentages of single- and multi-vehicle crashes as weights. Suggested Research This research identified the following areas in which addi- tional research is needed to address areas where the available data are too limited to support meaningful conclusions: 1. The effect of median barrier was identified in this research and in the literature. However, the small number of seg- ments with barriers did not allow for evaluation of the effects of barrier type or of the interaction between barrier presence and barrier width or barrier proximity to the travel lanes. These issues should be addressed in the future to determine the effectiveness of median barriers and to review existing guidelines for their placement. The literature suggests that barrier type can influence crashes; this is another area of potential future work. 2. An original goal of this research was to determine the effect of the number of lanes and lane width on crashes for multilane rural highways. The available data did not allow for estimating this effect. The effect of lane width has been documented in past research, and it was demonstrated to have an effect on crashes. This is a design element that could influence driver behavior and operating speeds and, therefore, additional attention should be paid to determin- ing the safety implications from lane width trade-offs. 3. The effects of paved shoulders and the presence of left-turn lanes were identified in this research, but pro- vided seemingly counter-intuitive results. The models showed that crashes increase on segments with paved shoulders and left-turn lanes. These features are generally C H A P T E R 5 Conclusions and Suggested Research
33 considered to be safety improvements; as such, additional research is needed to determine their effectiveness and to determine whether conditions exist where their presence may indeed contribute to crash occurrence. It is possi- ble that the presence of paved shoulders may encourage higher speeds, while the presence of left-turn lanes may create an obstacle in the road that has an impact on spe- cific crash types. 4. The research team discovered that there is a lack of uniformity among the various state databases that are available in the HSIS. Although the HSIS was developed to provide datasets that could be used in research to establish and evaluate nationwide trends, several variables are not common to all states. Further areas of concern are the differences in the level of detail provided by each state and the inconsistencies in the coding within common variables. For example, values of common variables are not coordinated; this often leads to aggregation of data to fewer detailed categories or even to binary (yes/no) values. Some effort is recommended to normalize these entries and to develop a truly uniform data set that will facilitate improved nationwide research evaluations.
