Penjadwalan Produksi menggunakan Algoritma Genetika dan Knowledge Based Approach pada PT Industri Pembungkus Internasional

Abstract

Production planning and operational scheduling very important for complexity of production system. The existence rules of delivery ontime, order arrival flexibility, differences of priorities, mixed production volumes, differences of machine capabilities cause uneven capacity utilization, job accumulation, and production delays. This study aims to handle parallel machine scheduling to get the best job sequence results to minimize job delays using genetic algorithms and knowledge based approaches (KBA). Genetic algorithms are one of the algorithms that have succeeded in finding solutions to optimization problems by applying the evolution process and eliminating bad solutions. KBA helps decision-making solve problems related to create a computing system imitates human intelligent behavior. The combination of the methods and earliest due date (EDD) rule produce an inference engine to build more adaptive initial population. The results of the proposed scheduling show the new rules successfully guide the search process more adaptive. The genetic operation process increases the fitness value from the initial initialization to the end of genetic operation process when the job is overloaded or underloaded. When overload, the fitness value increases by 0.95%, total number of job delays in initialization decreases from 198 jobs to 184 jobs and the average load capacity ratio (LCR) decreases by 7.54%. When the job underload, the fitness value increases by 3.56%, there are no job delays, and the LCR increases by 4.66%. The results of the proposed job scheduling sequence can meet the company's objectives by reducing lateness and increasing fitness values. The implementation of the integration of the two methods with the VB.Net programming language requires an average computing time of 32 seconds when running.