Optimization of Multi-Picking Route in Single-Aisle Operations Based on Forklift Modeling

This study presents a route optimization framework for manual forklifts in warehouse environments, utilizing regression-based travel time modeling alongside the ACOpt algorithm, which integrates Ant Colony Optimization (ACO) with K-opt techniques. Recognizing the significant impact of travel time on warehouse efficiency, a regression model was developed to accurately predict travel time based on key factors such as Euclidean distance, average height, and acceleration adjustments. This model refines route planning by enabling the selection of optimized paths for forklifts, with model predictions validated against actual travel times to confirm alignment. Experimental results based on one month’s worth of Work Orders (WO) demonstrate that ACOpt surpasses traditional Greedy and LKH algorithms, achieving a higher rate of route improvements while maintaining reasonable computational times. These findings underscore the potential of the proposed framework to enhance operational efficiency and suggest scalability to multi-aisle systems for broader warehouse applications.