Data-driven diagnosis framework for platform product supply chains under disruptions

Global supply chains face disruptions from geopolitical conflicts, pandemics, and wars. These disruptions exert a long-lasting effect across the supply chain, affecting supply, logistics, and markets. Platform product supply chains, characterised by their diversity of choices within interconnected nodes encompassing product configuration, supply, manufacturing, and delivery, are particularly vulnerable to these disruptions, incurring significant costs and diminished customer satisfaction. Therefore, the ability to diagnose these issues is vital for improving its overall performance. This study introduces a novel three-phase framework for supply chain diagnosis that leverages a data-driven methodology. Initially, the framework employs Generic Bills-of-Materials (GBOM) for qualitative structural mapping of platform products and their supply chains. Subsequently, a network model is constructed to encapsulate intra-nodal and inter-nodal dynamics of the supply chain. The third phase integrates Failure Mode and Effects Analysis (FMEA) with historical data to formalise supply chain domain knowledge, enabling a comprehensive analysis of the supply chain operational state. Finally, a real industrial case is presented, showing the effectiveness of the proposed framework in diagnosing short-, medium-, and long-term decisions. Findings reveal (i) inventory placement yield divergent impacts on the supply chain order fulfilment cycle time (OFCT) and (ii) reducing product variants improves planning accuracy and reduces OFCT.