Second-life lithium-ion batteries are increasingly considered for stationary energy storage and renewable integration. However, a major challenge in second-life deployment is the absence of reliable historical usage data. Key operating parameters such as cycle count and C-rate exposure are typically unknown, making accurate assessment of State of Health (SoH) and Remaining Useful Life (RUL) highly uncertain. This work presents a regression-based diagnostic–prognostic framework that reconstructs hidden operating history directly from measurable degradation indicators. Using experimental cycling data from LFP 18650 cells, multivariate linear regression models were developed to describe capacity fade (ΔSoH) and internal resistance growth (ΔRi) as functions of cycle number and C-rate. These continuous degradation surfaces form a data-driven of battery ageing. An inverse estimation procedure is then applied to the regression surfaces, enabling reconstruction of historical cycle count and C-rate from observed degradation data. Validation against known test conditions demonstrates history recovery with less than 2% relative error. The reconstructed state is subsequently used as the initial condition for scenario-based RUL prediction. Projection to a 70% SoH threshold shows that reducing discharge stress from 1.97C to 0.5C increases remaining lifetime from 1,912 to 2,579 cycles, corresponding to a 35% extension in second-life operation. Sensitivity analysis further indicates that charging at higher C-rate induces approximately 3.2% greater capacity degradation than equivalent discharging C-rate over 1,500 cycles, highlighting the dominant role of charging management in ageing behaviour. The proposed framework provides a transparent and data-efficient method for second-life battery assessment, linking measurable degradation directly to operating history reconstruction and future lifetime prediction.