A framework for multi-level decision support in manufacturing: A hierarchical and distributed approach

Manufacturing systems are increasingly exposed to high levels of complexity arising from volatile demand, tighter resource constraints, and the need to balance multiple, at times conflicting, performance objectives. Although Decision Support Systems (DSS) have been widely adopted to assist manufacturing decision-making, their implementation in industrial contexts remains largely fragmented. Existing DSS solutions provide limited guidance on structuring and integrating decision support across multiple decision levels, despite the growing digital maturity of manufacturing ecosystems. This paper proposes a hierarchical, distributed decision-support framework to address this gap. The main contribution of the paper lies in formalising a generic and modular coordination architecture for integrating diverse DSSs through vertical coordination, horizontal coordination, and structured feedback mechanisms. The framework enables the coordinated interaction of multiple DSSs across strategic, tactical, and operational decision levels while preserving local autonomy. It explicitly incorporates mechanisms for vertical and horizontal information exchange and structured feedback, supporting coherent and consistent decision-making in dynamic manufacturing environments. The applicability of the proposed framework is evaluated through a simulation-based manufacturing use case inspired by an automotive assembly environment, where a high-level production planning DSS coordinates three low-level DSSs for production scheduling, inventory replenishment, and maintenance planning. A Monte Carlo evaluation over five operating scenarios compares the proposed full framework against a siloed baseline and two partial coordination configurations. The results show that the full framework consistently improves system-level performance, increasing service levels while reducing final backlog, stockout-related disruptions, total operating cost, and constraint violations. In the most demanding combined-stress scenario, the framework does not eliminate disruption effects but achieves the best overall response, reducing final backlog and operating cost while avoiding stockout days observed in the siloed baseline. These findings indicate that hierarchical and distributed DSS coordination can improve the robustness, consistency, and operational feasibility of manufacturing decision-making under uncertain conditions.