Multiscale Production Routing in Industrial Gas Supply Chains

The production routing problem (PRP) considers the coordination of production, inventory, distribution, and routing decisions in a supply chain. In this work, we focus on production routing in the industrial gas business, where the challenge lies in the simultaneous optimization of complex and power-intensive production plants and highly integrated supply chains with vendor-managed inventory (VMI). We present a multiscale PRP (MPRP) model with two time grids, to which an MILP-based heuristic solution approach is applied to solve large-scale instances. The proposed MPRP framework is applied to a real-world industrial gas supply chain with 2 plants, approximately 240 customers, 20 vehicles, and a planning horizon of 4 weeks. The results show that the proposed solution method clearly outperforms available alternative approaches in terms of solution quality.