A comparative study on sampling strategies for truck destination choice model: Case of Seoul metropolitan area

One of the major issues when applying truck destination choice models with a large number of alternatives is how to sample a set of non-chosen traffic analysis zones (TAZs) to construct a destination choice set. Despite the large number of studies applying various sampling strategies, the question remains as to what are optimal strategies in model development. This study examined how the sampling strategies affect the performances of truck destination choice models. Two sampling methods (simple random sampling and stratified importance sampling) and four different sample sizes were tested using the truck trip data of Korea. For stratified importance sampling, Moran's I statistics were used to divide the entire study area into multiple strata, and Neyman allocation was used to determine the appropriate number of samples for each stratum. The truck trip productions were distributed by a Monte Carlo simulation, and two measurements of effectiveness (MOEs), average trip length (ATL) and trip length distribution (TLD), were used to evaluate and compare the performance of the destination choice models with respect to the sampling strategy. The results showed that the models using stratified importance sampling with smaller sample sizes performed better than others.