In the modern manufacturing and logistics landscape, balancing workforce costs with service quality remains a persistent challenge. Traditional integer programming (IP) approaches for manpower shift planning (MSP) often require advanced mathematical expertise and involve substantial computational effort, which limits their practical applicability in small and medium-sized enterprises.This study proposes a data-driven shift planning heuristic for a chemical logistics warehouse employing an atypical workforce. Instead of pursuing theoretically optimal but computationally intensive solutions, the proposed approach applies K-means clustering to identify representative demand patterns across work periods. These patterns are then used to guide the design of flexible shift start and end times.A real-world case study demonstrates that the clustering-based heuristic improves schedule flexibility, reduces overtime requirements, and supports compliance with labor regulation constraints. By aligning shift structures with demand density while accounting for practical workforce limitations, the proposed method generates feasible shift plans within short computational times. The results suggest that a simplified, clustering-driven heuristic provides a practical and efficient alternative to traditional optimization approaches in fast-paced manufacturing environments.

Keywords

Atypical Workforce, Manpower Shift Planning (MSP), K-means Clustering, Heuristic Algorithm.