Development of the hybrid BWM-fuzzy preference programming approach using a two-phase model: A case study in the automotive after-sales services

Multi-attribute decision-making (MADM) methods are essential tools for solving complex decision problems by offering systematic approaches for evaluating and ranking multiple criteria and alternatives. Despite considerable progress in MADM techniques, challenges persist in ensuring the reliability and consistency of outcomes, particularly under fuzzy environments. A recent advancement the hybrid best-worst method with fuzzy preference programming approach (the BWM-FPP model) has demonstrated promise, notably in hospital performance evaluations. However, the weight vector derived from the BWM-FPP model is a locally weakly efficient, and but not necessarily a locally efficient weight vector, which limits its robustness in practical applications. To address this shortcoming, we propose an enhanced two-phase model that guarantees the derivation of a locally efficient weight vector. In Phase ۱, the model maximizes the minimum membership degree, ensuring a strong baseline of reliability, while Phase ۲ improves overall robustness by maximizing the mean membership degree. The proposed model's effectiveness is demonstrated through theoretical validation, four numerical examples, and a real-world case study in the automotive after-sales service sector. Results show significant improvements in membership degrees and reliability, validating the model’s practical applicability. In the case study, key service quality criteria such as Reliability, Tangibles, and Assurance emerged as most influential, underscoring their critical role in after-sales service evaluation.

کلیدواژه ها:

BWM-FPP model ، Two-phase model ، Locally efficient weight vector ، Membership degree ، After-sales service criteria