Multi-class hazmat distribution network design with inventory and superimposed risks

Transportation and inventory are essential to hazardous materials logistics, while different classes of often transported over a network simultaneously. Despite their in-transit storage incompatibility, the superimposed risks among materials, which results from possible chemical reaction once accidents (e.g., leakage, explosion) happen, further complicate comprehensive plans. In this study, we introduce new multi-class hazmat distribution design problem with (MHND) in multi-echelon supply chain, where planning locations, inventory, routes made together. The long-term detour cost/risk cyclic time windows penalty costs under time-dependent (periodic) road closure policy explicitly formulated. We propose population-based risk definition that evaluates for population at any location respect its logistics system. particular, capture interactions between types superposition coefficients accommodate general system more than two types. develop knowledge-based NSGA-II algorithm dissimilarity-based elitist selection (NSGA-II-CD) solve problem. devised (CD) operator can tackle issue speeding proliferation, greatly improves solution quality. Our model is applied metropolitan-wide real-world case study Guangzhou, China. suggest that, perspective traffic management sector, periodic closures Guangzhou could be possibly upgraded full-time prohibition. Moreover, provide following insights authorities (1) there positive convex relation minimization equilibration. should not try find perfect risk, they make trade-off equity total exposed risk; (2) correlation level service risk. Thus, practice, agencies economic viability system, maintaining good customers; (3) considerably affect design; specifically, route overlapping ratio decreases when intensify.