Cost-effectiveness and tail-risk evaluation of remanufacturing modes for automotive drive axle housings

Selecting an appropriate organizational mode for automotive component remanufacturing remains challenging because process-quality uncertainty, cost-effectiveness, downside economic risk, and policy conditions are often evaluated separately, making it difficult to identify a mode that is both economically feasible and robust under adverse conditions. To address this problem, an integrated cost-effectiveness and tail-risk evaluation framework was developed for stamped-welded automotive drive axle housings using enterprise-level cost and process records together with damage statistics from 584 axle housings. Three organizational modes: OEM in-house remanufacturing, third-party outsourcing, and platform alliance were compared using AHP-TOPSIS. A 20,000-run Monte Carlo-CVaR analysis was then employed to quantify the uncertainty and lower-tail risk of the benefit-cost ratio R _c , followed by NPV evaluation under carbon-pricing and green-credit scenarios and sensitivity analysis of the non-remanufacturable ratio x and first-pass yield y . The results show that the platform alliance achieves the highest TOPSIS closeness coefficient ( C _i  = 0.81), compared with 0.72 for third-party outsourcing and 0.65 for OEM in-house remanufacturing. For the platform alliance, the mean R _c is 1.47, the 5% left-tail CVaR _5% is 1.02, and the probability of R _c  > 1 reaches 97.8%. Increasing the carbon price from 0 to 720 CNY/tCO ₂ e produces only a limited improvement in project NPV, whereas green credit increases the NPV of the platform alliance from CNY 3.61 million to CNY 4.10 million. Sensitivity analysis further identifies x  < 0.20 and y  > 0.85 as the critical operating window. These results indicate that the platform alliance provides the most favorable overall balance between economic performance and downside-risk robustness, while process-quality control and financing conditions exert greater leverage on project viability than carbon pricing alone. The proposed framework therefore provides an engineering-interpretable basis for remanufacturing mode selection, quality-threshold setting, and low-carbon financial-policy response.