An accelerated branch-and-cut algorithm for globalized robust sustainable biomass supply-chain optimization under uncertainty

Abstract

This article addresses the optimal design of a sustainable biomass supply chain under multiple sources of uncertainty, including fluctuating transportation costs and ambiguity in the demand distribution. A globalized robust bi-objective optimization model is proposed that integrates globalized robust optimization for transportation-cost uncertainty with ambiguous chance constraints based on a sub-Gaussian ambiguity set for demand. The model is reformulated as a tractable mixed-integer linear program through robust-counterpart reformulation. To improve computational efficiency, an accelerated branch-and-cut (A-BC) algorithm with problem-specific strengthening inequalities and a warm-start heuristic is developed. Computational experiments based on a Henan Province case with 102 supply points, 46 power plants and 102 candidate collection stations show that the Full A-BC configuration solves the standard test instances within the prescribed time limit and reduces solution time by 51.7% on average relative to the Gurobi-generic-cut baseline across the reported comparable instances. The optimized network yields total cost savings of approximately CNY 6 million compared with a conventional robust model and reduces transportation-related carbon emissions by 6.1% relative to a nominal deterministic model. The proposed approach provides an effective decision-support tool for the design of sustainable biomass supply chains under uncertainty.