Optimization of the Warehouse Logistics System, through the Application of Lean Warehouse and Machine Learning Algorithms

Abstract

Abstract. In 2022, the alcoholic beverage market in Peru experienced Growth due to the reduction of COVID-19 pandemic restrictions is pro- jected to reach pre-pandemic demand levels by 2026. In this context, logistics efficiency becomes crucial for profitability and customer satis- faction. This study proposes the combination of Lean Warehouse and Machine Learning algorithms to optimize the warehouse logistics sys- tem, it covers the preliminary analysis, implementation, and analysis of results, various Lean Warehouse techniques were applied, such as 5S, SLP, FEFO, and multicriteria ABC analysis, at the same time, Ma- chine Learning algorithms were applied, such as forecasting (SARIMA- LSTM), which allowed accurate forecasts of future demand and favored the distribution of the warehouse, as well as clustering (K-means) for the optimal grouping of products according to their expiration date. Key Performance Indicators (KPIs) were also introduced to gauge the success and efficiency of the logistics system. The results of the research showed substantial improvements in logistics efficiency, such as a reduc- tion in processing time per order guide by 103 minutes and an increase in process flow by 32.56%. These improvements benefited the company in terms of costs and efficiency, with a reduction in lost sales of S/34,386.75. Organizational adaptation and continuous management are essential to maintain and improve results over time. Keywords: Lean Warehouse · Machine Learning · Logistics optimiza- tion.