A two-phase production and transportation optimisation methodology for sustainable industrialisation under uncertainty

Achieving sustainable and resilient industrialisation requires integrated and adaptive decision-making across the entire supply chain, particularly amid uncertainty in demand, lead times, and production constraints. This study proposes a two-phase optimisation methodology that jointly addresses stochastic production scheduling and deterministic transportation planning in manufacturing environments. The novelty of this work lies in its sequential integration: the first phase develops a Mixed-Integer Linear Programming model enriched with Monte Carlo simulations (MCs) to generate optimal production quantities, subcontracting levels, and inventory decisions under realistic demand variability and fluctuating lead times. In the second phase, the output from the stochastic production model is used as input for a transportation optimisation problem. This phase employs Vogel’s Approximation Method (VAM) and the Modified Distribution Method (MODI) to construct a cost-effective and feasible transportation plan across multiple production and demand nodes. While existing methods often address production and transportation separately or assume deterministic inputs, this approach explicitly models production uncertainties. It links them to transportation outcomes, enabling more realistic and coordinated decision-making. This framework is validated using actual data from a cocoa manufacturing enterprise. The implementation yields an optimal production cost of GH¢ 2,600,454.25 and a transportation cost of GH¢ 1,212,394. Sensitivity analysis demonstrates that while the total cost is affected by small fluctuations in unit parameters, the transportation allocation remains stable, indicating the robustness of the solution. The proposed approach provides a practical and scalable tool for manufacturers. It enhances operational efficiency, reduces costs, and improves supply chain resilience.