Method production scheduling using a comparison of genetic algorithm and other general methods

Abstract

The goal of this study is to create the highest efficiency production work schedule. The followingmethods were used to design and develop the program to solve the problems : first in-first out (FIFO),shortest processing time (SPT), longest processing time (LPT), early due date (EDD), and minimizeslack. The researcher had developed the genetic algorithm (GA) method which is the holistic methodused for comparing the production scheduling outcome. Efficiency of the genetic algorithm wasexamined by designing a factorial 23 test for finding an appropriate parameter value using anovamethod. It was found that the appropriate value of the crossover, mutation, population size, andgeneration were equal to 1, 0.1, 100, and 50, respectively. This helped create a highest efficiency inthe management of work schedule.With regards to 4 problems in the management of work schedule testing of the mental alloyed factory,it was found that the GA and LPT methods could provide the best work value. Based on the examiningof problems in the production scheduling for 13 tasks of 3 task stations, problems in 23 task of 5 taskstations, and problems in 27 tasks of 4 task stations, the value finding was found at a lower bound. Inthe case of problems in 27 tasks of 5 task stations, the NP-Hard problem was found. The GA methodcould provide the best work value, making the mental alloyed factory has alternatives for the productionscheduling. Besides, it could be employed for an increase of the efficiency in the operation of thisfactory.