Enhancing Fisheries Sustainability Through Supply Chain Efficiency with Business Intelligence (Machine Learning) at Auction Sites

The fisheries industry faces complex challenges in supply chain efficiency that impact sector sustainability and the welfare of fishermen. This study aims to analyze the implementation of machine learning-based business intelligence systems to improve supply chain efficiency at Palang Fish Auction Place (TPI), Tuban Regency. The research method employs a mixed-methods approach, combining qualitative methods through in-depth interviews with fisheries stakeholders with quantitative methods using linear regression models to predict fish catch volumes for the 2022-2024 period. Qualitative data analysis employs the Miles & Huberman framework, which involves data reduction, data presentation, and conclusion drawing. In contrast, quantitative data is evaluated using metrics such as MAE, MAPE, and RMSE. The results reveal five primary factors influencing supply chain efficiency: catch volume with distinct seasonal patterns, auction price stability influenced by demand and import policies, distribution constraints resulting from inefficient payment systems, significant weather and environmental impacts, and the potential for technology adoption with positive acceptance among fishermen. The machine learning model successfully predicts catch volume with increasing accuracy from MAPE 18.5% (2022) to 12.8% (2024). The implementation of machine learning-based business intelligence systems has proven capable of improving fisheries supply chain efficiency, stabilizing fish prices, reducing resource waste, and supporting the sustainability of the fisheries sector in accordance with the Sustainable Development Goals. Keywords: Business intelligence, fish auction place, fisheries, machine learning, supply chain efficiency