Agent-based risk analysis model for road transportation of dangerous goods

Abstract

Dangerous Goods Transportation (DGT) is often referred to as hazardous materials or hazmat transportation. DGT is considered a critical aspect of modern logistics. It involves the movement of substances that are capable of posing significant societal and environmental risks for exposed populations, properties, and infrastructures in different environments. These substances, ranging from flammable liquids and explosives to toxic chemicals, play crucial roles in various industries, including manufacturing, healthcare, and research. Although essential for societal progress, transporting dangerous goods demands meticulous planning, stringent regulations, and specialized handling procedures to analyze potential hazards and asses shipment risk levels. In this paper, a DGT risk assessment method has been proposed and modeled using a multi-agents system, where each agent possesses specific attributes, failure modes, and behavioral patterns. Furthermore, the modeled system is simulated using two approaches. The first approach evaluates the overall risk level arising from a single DG expedition, while the second approach assesses the risk generated from a single accident using Monte Carlo simulation (MC). A Geographic Information System (GIS) was created to depict the risk analysis for both simulation methods. This system displays the location, trajectory of truck, and the affected areas in the event of an accident during simulation.

Keywords

• Multi-agent system
• Failure detection
• Risk analysis
• Agent-based modeling
• Monte Carlo simulation
• Stochastic systems