Genetic Crossover Operator in Local Search of the Assembly Line Balancing Problem

Abstract

One of the biggest challenges facing the industry today lies in the assignment of robots to the workstations of production processes in order to perform tasks collaboratively between robots and human labor; this is known as Cobots. This document proposes to minimize the number of workstations by assigning robots, to reduce production costs. Additionally, it seeks to minimize the number of workers by assigning Cobots to a percentage of workstations. Due to the above, in this research, the Assembly Line Balancing Problem Type I (SALBP-1) is adapted with a Cobot assignment problem. This problem is considered NP-hard, so the use of a hybrid heuristic is proposed in the search to minimize the number of workers and the number of workstations. The hybrid consists of a modified genetic algorithm applying a local search within the genetic selection operator. The results demonstrate a decrease in the assignment of workstations, as well as a decrease in assignment of workers when using Cobots in some workstations using the proposed algorithm for the benefit of organizations.