An intelligent decision-making system integrating fuzzy logic and machine learning for pharmaceutical closed-loop supply chain and production planning in an uncertain environment

Purpose This paper aims to propose a novel decision-making framework that integrates fuzzy logic with machine learning to tackle uncertainties in supply chain management and production planning. The focus is on optimizing inventory levels, balancing supply and demand and achieving production targets within a closed-loop supply chain (CLSC). Design/methodology/approach An intelligent algorithm is developed to monitor and adapt plant operations in real time using fuzzy inference and machine learning. The framework is validated through two case studies: an industrial hospital laundry (IHL) with 16,000 kg daily production, and a furniture company facing seasonal demand fluctuations. Performance is evaluated through inventory stability, production efficiency and cost savings, aligned with Industry 4.0 / 5.0 objectives. Findings The system effectively mitigates supply–demand imbalances, reducing inventory deviations by up to 28% and improving production schedule adherence by 21%. In the IHL case, fuel and detergent costs were reduced by 12–15% with an 18% rise in machine utilization. For the furniture company, forecasting accuracy improved by 24%, leading to 17% fewer stockouts and a 14% reduction in holding costs. Overall, the framework enhanced operational efficiency by nearly 20% while demonstrating strong adaptability under uncertain conditions. Originality/value By combining fuzzy logic and machine learning for real-time decision-making, this study contributes a scalable and intelligent solution for CLSC environments. Its quantifiable improvements in cost, efficiency and resilience underscore its relevance to sustainable, digitalized supply chains under the Industry 4.0 / 5.0 paradigm.