The development of safety performance functions (SPFs) and crash modification factors (CMFs) requires data on traffic exposure. The analysis of motorcycle crashes can be especially challenging in this regard because few jurisdictions collect motorcycle traffic volume data systematically. To address this challenge, the project team conducted several analyses to explore (1) how much predictive power for an SPF is lost when motorcycle volumes are unknown and how this lack of information may affect the development of CMFs for motorcycle crashes, and (2) alternative methods for deriving accurate predictions of motorcycle crashes or motorcycle volumes. The results of the analyses show that when motorcycle volumes are not known, using total average annual daily traffic (AADT) on its own is sufficient for developing SPFs and CMFs. The potential bias due to missing motorcycle-specific AADT is sufficiently negligible where it exists so as not to preclude SPF and CMF development. The project team also concluded that attempting to predict motorcycle volumes is not possible using typically available roadway and county-level data. Improvement could possibly be found in trip generation type modeling at a disaggregate scale, although given the success of SPF development using total AADT, such an effort may not be worthwhile. A more significant issue in developing motorcycle crash SPFs and CMFs is working with relatively rare crash types. In the analyses undertaken, SPFs could not be developed for all motorcycle crash types or site types. More evidently, in the estimation of CMFs using simulated data, the CMF value varied significantly between simulation runs due to the low frequency of motorcycle crashes. In terms of research gaps, a database is needed that includes implemented countermeasures expected to affect motorcycle crashes along with the location, date of treatment, and treatment description. This information would aid researchers in identifying treatments that are feasible for study. The report also identifies several research gaps related to analytical methods, related gaps, and data limitations.

The objective of the study was to evaluate the safety effects of two countermeasures with respect to vehicle–pedestrian crashes—the provision of protected or protected/permissive left-turn phasing and the provision of leading pedestrian intervals (LPIs)—using a before–after empirical Bayesian methodology. The study used data from North American cities that had installed one or both of the countermeasures of interest, including Chicago, IL; New York City, NY; Charlotte, NC; and Toronto, ON. This study showed that the provision of protected left-turn phasing reduced vehicle–vehicle injury crashes but did not produce statistically significant results for vehicle–pedestrian crashes overall. A disaggregate analysis of the effect of protected or protected/permissive left-turn phasing on vehicle–pedestrian crashes indicated that this strategy may be more beneficial when there are higher pedestrian and vehicle volumes, particularly above 5,500 pedestrians per day. At these high-volume locations, the left-turn phasing evaluation resulted in a potential benefit–cost (B/C) ratio range of 1:15.6::1:38.9. The evaluation of LPIs showed that the countermeasure reduced vehicle–pedestrian crashes. This evaluation produced a crash modification factor of 0.87 with a potential B/C ratio range of 1:207::1:517.

The Highway Safety Manual (HSM) is a tool that helps transportation agencies make data-driven decisions about safety. It includes methods for quantifying safety performance and predicting crash frequencies. The HSM is currently being updated to include macro-level crash prediction models, which can be used to assess safety trends at a regional or national level. NCHRP Research Report 1044: Development and Application of Quantitative Macro-Level Safety Prediction Models, from TRB's National Cooperative Highway Research Program, details macro-level models, which have the potential to be a valuable complement to micro-level models, which are currently the only type of model included in the HSM.