Clustering and Network Analysis of Mobility Patterns as an Analysis Tool for Lean Project

The study aims to optimize internal logistics processes by applying Lean philosophy and data science tools, with a primary focus on qualifying processes to determine their value-added contribution within the logistics context. Utilizing a novel two-step methodology, the research first employs a modified DBSCAN algorithm to analyze indoor positioning data and categorize activities. This is followed by multi-layer network modeling to understand processes and create a framework that enables the reduction of idle activities through optimization algorithms. A real warehouse case study, using a UWB-based Indoor Positioning System (IPS) to track forklifts, demonstrates the method's effectiveness in identifying non-value-added activities. The results reveal specific opportunities for reducing idle, enhancing resource utilization, and improving operational efficiency. This innovative combination of advanced data analysis techniques and Lean principles provides a comprehensive framework for logistics optimization, significantly enhancing process efficiency through optimized task scheduling and resource allocation. Doi: 10.28991/ESJ-2025-09-01-013 Full Text: PDF