A multi-criteria supplier evaluation and selection model without reducing the level of optimality

Traditional mathematical models for supplier evaluation and selection have primarily focused on cost-oriented perspectives, neglecting non-cost factors such as strategic and relational measures. However, such models have often compromised the optimality of solutions. Additionally, the use of genetic algorithms in these models has resulted in similar objective function outputs with minimal differences, making it challenging to differentiate between potential suppliers. To address these limitations, this study proposes a novel approach that employs multi-criteria decision-making (MCDM) techniques to evaluate suppliers without sacrificing optimality or imposing additional constraints. The proposed model utilizes hybrid algorithms and incorporates the weighted criteria obtained through the Best–Worst Method and the ranking and selection of alternatives using the Technique for Order of Preference by Similarity to Ideal Solution. By applying MCDM methods in the post-evaluation phase, this research introduces a unique approach that preserves solution optimality. The findings highlight that while the objective function outputs from the genetic algorithm implementations may be nearly identical, they can vary significantly across other criteria. This study demonstrates the strength of multi-criteria evaluation in overcoming the limitations of genetic algorithms without compromising solution optimality.