Research Information System
Supply Chain Analytics, vol.14, 2026 (ESCI, Scopus)
The rapid growth in global electric vehicle (EV) adoption has intensified the need for effective end-of-life battery management, positioning it as a complex supply chain analytics problem that connects sustainable energy, the circular economy, and infrastructure planning. This study develops an artificial intelligence (AI)–enabled decision support system that integrates four decision layers: prediction of battery retirement timelines, machine learning (ML)–based health diagnostics, chemistry-specific routing, and reverse logistics network optimization. Focusing on the Turkish market, the framework combines a logistic growth model to estimate future battery flows, chemistry-dependent retirement profiles for nickel–manganese–cobalt (NMC), lithium iron phosphate (LFP), and nickel–cobalt–aluminum (NCA) batteries, and an ML-based classification model that assigns batteries to reuse, recycling, or disposal pathways based on state of health (SoH) thresholds; a p-median facility location model is then used to design a cost-efficient reverse logistics network. Results indicate a two-wave battery retirement pattern beginning in the 2030 s, with an initial wave (2031–2036) dominated by NMC batteries and a later wave (post-2038) led by LFP batteries; of the projected 8.2 million batteries retiring by 2050, approximately 5.3 million are estimated to be suitable for second-life applications in stationary energy storage. The model further identifies a four-facility reverse logistics network as the most cost-effective configuration. These findings are based on scenario analysis rather than deterministic forecasts, and the main contribution is a data-driven, integrated planning framework for battery end-of-life management that combines forecasting, classification, and network design in an emerging EV market.