Forecasting Crashes in Urban Districts using Aggregate Crash Prediction Models

Safety conscious transportation planning is a new strategy in proactive approaches towards crashes. This approach requires specialized tools at macro-level to account for the safety implications of transportation policies and plans at an aggregate level. The purpose of his paper is to develop Aggregate (macro) Crash Prediction Models (ACPMs) that are concurrent with the trip generation step of the four step demand modeling technique. The concept of Crash Generation Model (CGM) is introduced based on trip frequencies generated by purpose in each urban Traffic Analysis Zone (TAZ) utilizing a logarithmic transformation in a generalized linear model with the assumption of a negative binomial error structure. CGMs may be used for immediate checking of the impact of future trip generations on crash frequencies in comprehensive transportation planning studies (i.e. ability to forecast crashes at each time-step trips are being forecasted). They will help safety planners develop travel demand management scenarios and simultaneously assess their perceived impact on the overall safety of the urban areas. The models prove an effective step towards incorporating safety into long-range transportation planning. It is shown that there is a significant relation between crash frequencies and trip frequencies by purpose per TAZ.