An Estimation of Distribution Algorithm-Based Approach for the Order Batching Problem: An Experimental Study

An Estimation of Distribution Algorithm-Based Approach for the Order Batching Problem: An Experimental Study

Ricardo Pérez-Rodríguez (Center for Research in Mathematics, Mexico) and Arturo Hernández-Aguirre (Center for Research in Mathematics, Mexico)
DOI: 10.4018/978-1-4666-9779-9.ch026
OnDemand PDF Download:
List Price: $37.50


In the supply chain and the planning and control of warehouse processes, the order picking is an aspect critical. Combining customer orders into picking orders to minimize the picking time is known such order batching. Extensive evolutionary algorithms haven been proposed to build better batches for the order picking. The authors think that any algorithm should preserve batches that appear frequently in all members of the population in order to keep track and inherit these characteristics exhibited by the parents to the next generation. However, the traditional evolutionary operators used in current research sometimes lose the characteristics mentioned. In order to describe the characteristics exhibited by the parents as a distribution of the solution space, the authors build a probability model. An acceptable performance using the model proposed is shown against different evolutionary algorithms known in the literature in a series of extensive numerical experiments.
Chapter Preview

Literature Review

A discussion about the most current research on the order batching process is outlined below.

A practical batching problem where greeting cards are retrieved from a warehouse was analyzed by Kamin (1998). In that environment, the pickers use automated guided vehicles on a fixed course collecting the items according to given customer orders. The Kamin’s research focuses on the minimization of average turnover times where the orders arrive throughout the study horizon.

The optimal number of customer orders that should be assigned to a batch such that the average turnover time is minimized is focused on Chew et al. (1999). They employ a queuing network with two queues. In the first queue, customer orders arrive according to a Poisson process and batches are generated by means of the FCFS rule (First Come First Serve). If a particular number of customer orders are in the first queue, those orders are assigned to a batch and move onto the second queue. Those orders are released according to the availability of pickers.

Key Terms in this Chapter

Online Optimization: The method for finding the best decision for online batching.

Probability Matrix: The layout of the probability model.

Order Batching: It consists in grouping different orders in only one batch.

Probability Model: The representation of the solution to describe the distribution of the solution space.

Pickers: Operators in a warehouse.

Estimation of Distribution Algorithm: A group of evolutionary algorithms based on probabilistic models to solve combinatorial optimization problems.

Warehouse Management: It improves the performance of the warehouse.

Order Picking: It collects different articles in a warehouse environment.

Complete Chapter List

Search this Book: