Comparison of Calibration Factors Obtained Using Oregon and Georgia Data to Implement HSM Procedures on Two-Way Two-Lane Rural Roads

A decade long research effort in the field of highway safety has resulted in release of the Highway Safety Manual (HSM), which provides tools for performing quantitative safety analyses. Predictive methods described in HSM help in identifying sites with potential for safety improvement (PSI). Considering two-way two-lane rural roads, the default safety performance functions (SPFs) used in HSM were generated using Minnesota and Washington data. When applying the HSM procedures in other states, calibration should be conducted to obtain the most reliable predictions for the local area. Some states may predict fewer, while others may predict more than those predicted using default SPFs. These calibration factors account for the variations in traffic patterns, climate, topology, etc. For this research, Oregon and Georgia states' data were used to calculate the respective calibration factors to perform Empirical Bayes analysis on two-way two-lane rural roads. These two states were considered as they are geographically on opposite sides of the spectrum, and both states have been early adopters of the HSM. The calibration factors and crash patterns were compared to understand the adaptability of default SPFs by various states and even within each state.