Minimization of Workstations on Assembly Lines Through the Adaptation of Genetic Algorithms

Abstract

The problem of balancing assembly lines frequently appears in organizations dedicated to transformation processes and services. The objective is to minimize the number of workstations to minimize production costs. The problem is classified as NP complete due to the computational complexity generated by the exponential growth in the number of solutions. Additionally, the problem becomes more complex due to the size of the product and the operator's possibility of performing simultaneous tasks. This is known as the multi-manned problem for type one assembly line balancing; because it seeks to minimize the number of workstations with a predetermined cycle time. The use of a genetic algorithm is proposed, the model was coded in such a way that the random solutions in the genetic operators are all feasible, which allows for improving the efficiency of the algorithm instead of generating random solutions, this is another of the innovations of the work. The results show that it is possible to reduce the number of workstations by assigning tasks in a balanced manner.