A Bayesian Best-Worst Method framework for analyzing waste generation factors in bicycle manufacturing industry

Waste reduction has become a strategic priority for improving sustainability and competitiveness in manufacturing industries. Despite the economic importance of Bangladesh’s export-oriented bicycle manufacturing sector, there is a lack of systematic investigation into the key drivers of waste generation. This study addresses the central research question: which factors most significantly contribute to waste generation in bicycle manufacturing, and how can they be prioritized for effective intervention? To answer this, a Bayesian Best-Worst Method (BBWM) was employed, supported by structured expert elicitation. In the absence of prior comprehensive frameworks, seventeen waste-generating factors were identified and classified into five clusters: process and quality control, planning and management, workforce and training, material and compliance, and maintenance and resources. The results indicate that process and quality control is the most influential cluster (0.265), followed by planning and management (0.242). At the global level, callousness in preproduction sampling (0.1036) ranks as the most critical factor, followed by insufficient quality control checkpoints and unclear work instructions. Additional high-impact drivers include centralized decision-making, high worker turnover, inadequate maintenance planning, and poor-quality raw materials. These findings demonstrate that waste generation is predominantly driven by deficiencies in operational control, quality assurance, and interdepartmental coordination. This study contributes a novel BBWM-based prioritization framework and provides actionable insights for practitioners. By targeting high-impact factors, manufacturers can implement focused strategies to reduce waste, improve process stability, and enhance the sustainability and global competitiveness of bicycle manufacturing systems.