Design of an Order Picking Reduce Module Using Bat Algorithm

Design of an Order Picking Reduce Module Using Bat Algorithm

Rogelio Florencia (Autonomous University of Juarez City, Mexico), Julia Patricia Sanchez-Solis (Autonomous University of Juarez City, Mexico), Ivan Carvajal (Autonomous University of Juarez City, Mexico) and Vicente García (Autonomous University of Juarez City, Mexico)
DOI: 10.4018/978-1-5225-8131-4.ch011


These days the human factor is key to improving order picking processes. For example, in a line of processes to replace different parts of equipment or devices, it is necessary to find the best route in each case to minimize the time consumed. One of the tools used to search for solutions are the metaheuristic methods that use probability to find the best solution in the required time. One of them is the “bat algorithm” that has had outstanding results since its appearance in 2010. This metaheuristic is based on how bats hunt, supported by their system of echolocation. It is one of the most recent metaheuristics and has proven to be more effective than other similar optimization algorithms in different processes. However, the algorithm was initially designed for addressing continuous problems. In this chapter, the authors present an adaptation of bat algorithms to solve a discrete problem: order picking.
Chapter Preview


The order picking is the process of classifying and pulling items in a warehouse before delivering them to customers. The order picking operations are the most important processes in a warehouse. The cost of these operations is estimated to be as much as 50% of the total warehouse operating expense (de Koster, Le-Duc & Roodbergen, 2007). According to Cano, Correa-Espinal and Gómez-Montoya (2018), the order picking operations are order batching, batch sequencing, picker routing, joint order batching and sequencing, and routing problem.

The picker routing problem consists in determining the best picking sequence to retrieve the items from their storage locations in order to satisfy a customer demand. To optimize the process, it is necessary to find the best route; the problem is that there are different models and the best route for each model is different. To solve these problems, there are different strategies as shown by Ai-Lan (2018) in his article, where he makes a comparison of different strategies and methods that are used to solve the problem of order picking and which strategy is better in each case. However, carrying out a similar process in each case and for each change of route can be a long and expensive process.

In order to find a solution for the picker routing problem, it could be transformed into a traditional Traveling Salesman Problem (TSP) with some considerations in a similar way as was done by Hu and Raidl (2008), transforming the Railway Traveling Salesman Problem into a classic TSP. The TSP is a widely studied combinatorial optimization problem which consists of a salesman who travels through several cities and must visit each one of them only once and finally return home. The challenge of the TSP is minimizing the total distance traveled by the salesman.

A metaheuristic is a search algorithm that seeks to find the best solution to an optimization problem. Regularly, metaheuristics are based on the behavior of a living being to solve a problem. It is designed to solve a problem in a general way, given certain parameters, for optimizing a problem. The Bat Algorithm is used for continuous problems. It is inspired by the way in which bats hunt their prey by means of their echolocation. Most bats use an echolocation system that is very efficient at the time of flight and when hunting. Echolocation means that they can locate their prey through the sound waves they emit, and the resulting echo allows the bat to find the position of the object or prey that it is looking for; human beings are unable to hear the ultrasonic waves emitted by them. This is like the echo that occurs when speaking inside a cave. This is how bats can hunt at night in the real world. Bats have several anatomical adaptations that allow echolocation, these adaptations are in their larynx and their ears, and the way they process these signals to locate. Therefore, this adaptation of the bats allows them to locate themselves without having to use sight like most species. Bats emit ultrasonic sounds from their nose through their larynx. The frequencies they emit are usually between 1,400 and 100 thousand Hertz, however, the human being has an audible range of 20 to 20 thousand Hertz, therefore, the being human is unable to hear these sounds, however the exact range in which bats emit these sounds can vary by species. The ears of bats have a complex shape that helps them to capture sound waves and contain numerous receptors sensitive to low frequencies. The behavior of bats has been widely studied by researchers and is used in devices today as in ship radars, but a metaheuristic has also been created that takes advantage of this method as seen in Figure 1.

In this chapter, a discrete version of Bat algorithm was implemented to address the picker routing problem. The implementation was based on the continuous algorithm developed in MATLAB by Yang(2010). The discrete version of this algorithm was based on the modification proposed by Jiang (2016). In order to evaluate the performance of the algorithm described in this chapter, test instances for order picking problems already existing in the Internet were used.

Figure 1.

Bat algorithm representation


Complete Chapter List

Search this Book: