Impact of the method of criteria normalisation on the order picking route and time

Purpose: The purpose of the article is the selection of the best criteria for the normalisation method in order to achieve the minimal order picking route and time.

Design/Methodology/Approach: When a company utilizes warehouse with the shared storage system, every product can be stored in many, sometimes very distant from each other, locations. Locations were selected by the multiple-criteria decision-making technique – TOPSIS. The research was conducted by means of the simulation methods. Every location was described by three criteria (distance from the I/O point, degree of demand satisfaction and the number of other picked products in the proximity of the analyzed location), for which 37 combinations of weights and 18 normalization formulas were applied. For every combination of weights and normalization method 1000 orders were generated.

Findings: The best results were obtained when high weight was assigned to the degree of demand satisfaction. It was hard to indicate unequivocally the best normalization method. However, quotient inversions generally yielded slightly worse results than standard scores and feature scaling.

Practical Implications: Obtained results indicate that the presented approach can be useful in real warehouse management. It is a quite versatile method that can be adopted to various situations.

Originality/value: Although the routing problem in order picking has already been widely discussed, the literature about the problem of selection of locations in shared storage is quite scarce. Therefore, the presented approach can serve as one method of selection of locations in the shared storage system.