Click for next page ( 28


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 27
27 CHAPTER 4 Design Elements Recommendations Chapter 3 described the development of prediction equa- Average Shoulder Width tions from which a set of recommended AMFs could be deter- mined for practicing engineers to use when evaluating safety Recommendation trade-offs for selected design elements. In this chapter, each The research team reviewed past literature, the recom- of the design elements that were found as statistically signifi- mended values for the HSM, and the AMFs from NCHRP cant in the prediction models is discussed with a view toward Project 15-27 and agreed that there is an influence on crash developing these practical recommendations. A general dis- occurrence from the presence of shoulders. Using this back- cussion of the literature is first presented (review Chapter 2 for ground information, the team determined that the values noted more detail). The values from the models developed in this for all crashes for undivided highways are reasonable and in research are then presented with a brief discussion demon- accordance with current rends and literature. The team fur- strating the appropriate AMF values. Finally, a recommended ther recommended the use of only the AMF for all crashes for set of values is presented along with discussion of the justifi- undivided highways since the shoulder width was not a sig- cation for the proposed AMFs. The justification and associated nificant variable in the single-vehicle models. discussion is considered essential, especially for cases in which The team considered the values provided for all three models the results seemed to be inconclusive. for divided highways, and it recommended using the values The first step in establishing the proposed recommended from the single-vehicle crashes as those for divided roadways. values for each design element was the circulation of a draft set The team determined that the values for multi-vehicles and of recommendations among project team members. The team all crashes were high and probably reflective of other influences then met to discuss (1) the proposed values, (2) the justification such as volume. This adjustment is considered justifiable based of the recommendations, and (3) the identification of any issues on previous work by Harwood et al. (26) and the recommended that might result in diminishing the practicality of the proposed values in the HSM (22). The recommended values are sum- values. The team meeting represented an expert panel since it marized in Table 16. included three safety engineers, two highway designers, and a The modification factors in Table 16 are for all crashes and highway safety analyst (see Table 15). The team debated the not for specific types of crashes that could relate to shoulder values presented, discussed the existing work (both past and width issues. The recommended values are similar to those that of NCHRP Project 15-27), and prepared a recommended proposed in the HSM as noted above, and those of the divided set of AMF values through a consensus-building process. highways are comparable for almost all categories with the only The team determined that the current recommendations for exception being that of the 8-ft shoulder AMF. For undivided shoulder width and median width were reasonable (i.e., the highways, the differences between the NCHRP Project 15-27 values were similar to those in past research, and any differences and the HSM-recommended AMFs were larger. These dif- in magnitude could be explained). The team was not able to ferences are attributed to the fact that the HSM factors were make a final recommendation for median barrier presence developed for shoulder-related crashes while the AMFs for because it was not included in all models for divided highways. NCHRP Project 15-27 were developed for all crashes. Even Finally, the AMFs developed for the presence of paved right though a comparison with the HSM values is not wholly shoulders and left-turn lanes produced counterintuitive results, appropriate due to the difference in crash types used in each and the research team concluded that neither AMF should be model, the comparison is supported by the observed similar- included as a design element with a guideline. ities in trends and agreement of findings. Future research