Analysis of production line balancing using genetic algorithms

Abstract

XYZ company has a problem on the production line that is the build up of work stations and bottlenecks that indicate an imbalance. The aim is to solve the problem of balancing the production line at the company. The production line using the Genetic Algorithm method is done by determining objective parameters and functions, making the encoding of the work station, determining the initial initialization according to the actual production path, then doing iteration with selection, crossover, and mutation to form a new population, and ending with the termination of the algorithm. The results of the actual production line obtained are the number of work stations as many as 8 work stations and with the largest cycle time of 2213 seconds. The genetic algorithm production trajectory is obtained after achieving the maximum objective function value or until the maximum iteration limit is reached and consists of 7 work stations with the largest cycle time of 2203 seconds, and has a higher efficiency value thereby minimizing idle time, having a higher value of balance delay low so that it shows a decrease in waiting time and a lower value of the smoothing index compared to the actual production line where the production line is more balanced, meaning that the division of work elements is quite evenly distributed on the assembly line.