Lag-Based Forecasting Model for the Availability of Second-Life Electric Vehicle Batteries in Energy Applications

The growth of electric vehicle markets creates a delayed supply of batteries for stationary energy storage. This study develops a lag-based cohort model for forecasting the residual capacity of EV batteries potentially available for energy applications. Vehicles entering a national market as new or used are assigned different remaining traction-service durations. Retirement is separated from collection, technical qualification, and allocation to energy reuse through conditional transition fractions, while residual capacity describes the selected subset. The Ukrainian application combines reported passenger BEV registrations for 2013–2023 with rounded passenger-entry reports for 2024–2025; observed, proxy, and assumed market shares are distinguished. Under the baseline scenario, annual capacity inflow is 0.197 GWh in 2025, peaks at 2.242 GWh in 2030, and reaches 1.201 GWh in 2040. Cumulative inflow over 2025–2040 is 13.439 GWh, with combined scenarios spanning 7.696–21.750 GWh. A common nine-year lag reduces cumulative inflow through 2030 from 4.865 to 0.900 GWh; an illustrative distributed lag lowers the 2030 peak to 1.515 GWh. Data-source and market-share sensitivities quantify dependence on entry accounting. Chemistry, condition, thermal history, and repurposing architecture are mapped to the retirement and selection parameters. The results distinguish retirement volumes from potential storage-capacity supply and support planning the timing and scale of collection, screening, and repurposing infrastructure. The framework links vehicle-entry cohorts to the timing and scale of reusable storage-capacity supply, providing a transferable basis for secondary-resource planning in markets with substantial used-vehicle imports.