Abstract:

In order to investigate the spatiotemporal instability on pedestrian accident injury severities， a random parameters approach with heterogeneity in means and variances was constructed to analyze the heterogeneity of pedestrian accident injury severities. Using the pedestrian traffic accident data， 21 influencing factors were selected， the temporal instability and spatial transferability of the model were assessed using the log-likelihood ratio test， while the average elasticity coefficient was used to analyze the influence of each factor on three types of accidents： severe injury， minor injury and no injury. The results revealed that the influencing factors of pedestrian accident severities have significant spatiotemporal instability. The influence of the four factors， namely driver escape， physical separation， dirt shoulder and signal control on the severity of the accident varies with the positive and negative of the year， and there was temporal instability in the three years. The influence of pedestrian crossing the road and dark lights at night on severe injury accidents varies positively or negatively across different regions， and there was spatial instability in urban and rural areas. The effects of improper pedestrian behavior and nighttime darkness with lighting on pedestrian accident injury severities vary positively or negatively with both year and region， displaying significant spatiotemporal instability. Moreover， the random parameters， as well as the heterogeneity in means and variances， also exhibit significant spatiotemporal instability in urban and rural areas across different years.