Modeling and Optimization of a Sustainable Blood Supply Chain Network under Disruptive Conditions Using LSTM-Based Fuzzy Demand Forecasting

Abstract

The blood supply chain is considered one of the most complex supply chains in the healthcare system due to the critical nature of blood products, their perishability, limited shelf life, demand uncertainty, and the potential occurrence of disruptions. Effective management of this supply chain requires approaches that can simultaneously ensure accurate demand forecasting, network resilience, and economic, social, and environmental sustainability. This study proposes an integrated forecasting–optimization framework for modeling and optimizing a sustainable blood supply chain network under disruptive conditions. In the first stage, the demand for blood products is forecasted under three scenarios using a Long Short-Term Memory (LSTM) neural network. To capture demand uncertainty, the forecasting outputs are transformed into fuzzy numbers and incorporated into the decision-making model. Subsequently, a multi-objective mathematical programming model is developed to design a sustainable blood supply chain network by simultaneously optimizing economic, social, and environmental objectives. To enhance the robustness of decision-making against parameter uncertainty and supply chain disruptions, a fuzzy robust optimization approach is employed. Furthermore, the proposed model is solved using the Non-dominated Sorting Genetic Algorithm II (NSGA-II), and sensitivity analysis is conducted to evaluate the model's performance under different disruption scenarios. The results demonstrate that the proposed framework significantly improves demand forecasting accuracy while enhancing network resilience, reducing total operational costs, mitigating environmental impacts, and improving service levels under disruptive conditions. Therefore, the proposed model can serve as an effective decision-support tool for blood service organizations and healthcare policymakers in designing sustainable and disruption-resilient blood supply chain networks.