A data-driven BWM-ANN hybrid approach for evaluating sustainable supplier selection

Gaining sustainability is one of the aims of most of the manufacturing industries now-a-days. The suppliers have key roles in the achievement of this sustainability. This is because inadequate knowledge of the suppliers about the implications of sustainability can cause difficulty proceeding with sustainability achievement by the manufacturing industries also. Sustainable manufacturing industries are required for their own existence in future in a competitive challenging situation i.e., for the existence of workforce, product, process and customer. Therefore, the main purpose of the present study is to select the best supplier who can significantly contribute sustainability achievement to the manufacturing firms. Real life data from manufacturing companies have been collected for the present study and subsequently data analysis has been implemented to achieve the goal. The four components of data analytics viz. diagnostics, deterministics, predictive and prescriptive have been applied sequentially with the processed data. Finally, the computational technique selects the best supplier among the four considering the sustainable criteria. The prescriptive analytics` finally suggests the priority selection of suppliers for the industries. Along with this, Best Worst Method (BWM) and Analytical Neural Network (ANN) have been utilized for result analysis and validation. The results imply that data analytics significantly improves the accuracy and efficiency of supplier selection, ensuring alignment with sustainability goals. This study offers practical implications for manufacturing industries by demonstrating how data analytics can enhance sustainable supplier selection. The novelty of this research lies in its structured application of advanced analytics and provide sustainable criteria selection for sustainability-focused supplier evaluation, bridging a critical gap in current manufacturing practices.