Truck Trip Assignment to Utah's Truck Routes Using Geographic Information System and Genetic Algorithm

摘要

A truck trip assignment model for planning purposes was developed using a commodity-based model and a variant of the four-step model.The model uses a geographic information system (GIS) and a genetic algorithm (GA) to assign truck trips to truck routes.Truck trip production and attraction were estimated prior to the assignment step,using the results of a commodity-flow based trip distribution method.GA was used to find the optimal assignment of truck trips to Utah's truck routes by comparing RMSE of the truck counts from ATR stations with the truck trips assigned by a GIS software program.After running 81 generations consisting of 1,000 chromosomes through the GA,it was apparent that truck trips were more realistically assigned to the state’s truck routes than by the initial shortest path assignment.However,the differences between individual daily truck traffic as reported by ATR and the truck traffic estimated by the assignment model using GA were significantly larger than expected at some locations and further research is recommended.