A Hybrid Agent-Based Simulation and Optimization Approach for Statewide Truck Parking Capacity Expansion

Abstract

Abstract Truck parking shortages are a critical concern for both the trucking industry and truck drivers in the USA. To tackle growing shortages, parking capacity expansions are needed. This paper presents a hybrid agent-based simulation and optimization approach to model truck movements and driver behaviors to determine feasible locations for truck parking facility capacity expansions across a state. The simulation model considers the driving limit and rest requirements of hours-of-service regulations set forth by the federal regulations. By leveraging observed historical truck GPS data, agent profiles are created, capturing truck trip origins, necessary stops, trip destinations, and parking behaviors. The simulation model estimates parking facility utilization over time by identifying where, when, and how long truck drivers rest. The estimated parking usage data is then fed into a maximal coverage capacitated multiple facility location optimization with multiple service model to deduce capacity expansion locations given budgetary restrictions. Ultimately the model recommends where and how many parking spaces to add to accomplish a certain level of parking overcrowding. The simulation and optimization approaches are finally integrated into a map-based, user adaptable decision support tool.