Making Warehouse Logistics Smart by Effective Placement Strategy Based on Genetic Algorithms

Abstract

Supply chain executives are faced with the challenge of reducing labor costs. Travel time or picking efficiency can easily account for 50% or more of order picking time. If we omit human factor and the technical equipment of the warehouses, picking efficiency is mostly affected by two factors: correct combining orders into a single travel instance and picking orders in batch is the first factor; the second one is a goods placement – the more effective the goods are located, the shorter will be the picking distance for each order or batch of orders. It means that individual orders will be picked faster. Usually to determine the correct location for the goods 3PL’s are using ABC analysis that includes indicators like count of orders, goods turnover, picking rate, weight etc. There are also more complicated indicators like goods adjacency. Such indicators are harder to take into account using ABC analysis, as it requires sophisticated analysis of customer orders. In recent publication goods placing by results of ABC analysis was compared to the genetic algorithm approach. It was showed that genetic algorithm much more effective for goods placing. The goal of this paper is to improve developed genetic algorithm and include in calculations factors of the labor costs and warehouse topology. These factors will make algorithm usable in real warehouses and WMS (warehouse management system) information systems.

Optimization of Warehouse Operations with Genetic Algorithms

“…As the solution presented by Avdeikins and Savrasovs [2] may at the first glance seem similar to our solution, it is worth pointing out the differences. In our solution the fitness is calculated as a sum of all order picking route lengths.…”

“…Avdeikins and Savrasovs [2] applied genetic algorithms to warehouse optimization using order crossover (OX). Each individual in the population represented a warehouse layout.…”

“…Although this issue has been analyzed for a long time, and especially in recent years its intensive development has been taking place [2][3][4], we were not able to find in the literature a complete, accurate, fully automatic solution, which would consider the real distribution of orders in the optimization of product placements. All the papers we were able to find presented only some partial approaches, with big simplifications, for example that the route length is determined only by one product with the longest distance from the entrance [2], or that the user is responsible for re-allocating the products by placing the more frequently used closer to the entrance [5].…”

Hybrid Genetic Algorithms for Order Assignment and Batching in Picking System: A Systematic Literature Review

“…The study in [18] explored the challenge of reducing labor costs in supply chain operations, specifically concentrating on improving picking efficiency through precise order combining and strategic placement of goods. While ABC analysis is a traditional method for determining goods placement, recent research suggests that an approach involving genetic algorithms is more effective.…”