Applying a Home-Based Approach to the Understanding Distribution of Economic Impacts of Traffic Crashes

Abstract

The current practice of road safety attributes traffic crash costs to the location of traffic crashes. Therefore it is challenging to estimate the economic cost of traffic crashes and individuals who are more prone to the burden of traffic crashes. To address this limitation, this study used the home address of individuals who were involved in traffic crashes in the Knoxville Regional Travel Model (KRTM) region between 2015 and 2016. After geocoding the home addresses, 110,312 individuals were assigned to the Traffic Analysis Zone (TAZ) corresponding to their home address and the economic cost of traffic crashes per capita (ECCPC) was calculated for each TAZ. The average ECCPC in the study area was $1,250. The KRTM output was used for extracting travel behavior data elements for modeling ECCPC at the zonal level. This study also established an index to measure average zonal activity in the transportation system for each TAZ. Analysis indicates that the burden of traffic crashes was more tangible in the TAZs with lower-income households and higher average zonal activities. To account for spatial autocorrelation, a Spatial Autoregressive model (SAR) and a spatial error model (SEM) were used. The SAR model was more suitable compared with SEM and ordinary least squares regression. Findings indicate that average zonal activity and traffic exposure have a significant positive association with ECCPC. The ECCPC could be used as an index for allocating proper countermeasures and interventions to groups and areas where the burden of traffic crashes is more tangible.