Optimization of an RMG/Textile Plant through Discrete Event Simulation and Multi Objective Genetic Algorithm: A Case Study

Abstract

The RMG and textile apparel industry are key contributors to a country's economic growth. Bangladesh ranks second among the top textile-exporting countries. Despite this achievement, the supply chain of an RMG faces various obstacles, including increased lead time, excess inventory, transportation, and maintenance costs. Additionally, there are also the problems of the inefficient utilization of resources and overworking of manpower. Moreover, even a small percentage of inefficiency can cause a loss of thousands of dollars. This study aims to create a digital twin (DT) of a real RMG plant and optimize production parameters such as throughput, lead time, bottleneck, and resource utilization. At first, we collected data from a plant and used this data to create the simulation model using a Discrete Event Simulation (DES) software. After validation of the simulated model against real factory data, we used the Multi-Objective Genetic Algorithm (MOGA) to optimize the whole production line. Through our experiment, we found that a bottleneck in an operation was causing the throughput to decrease. After the implementation of MOGA and task-clustering method, we saw a significant productivity improvement. Thus, we can conclude that by the use of advanced technologies like DES and other optimization algorithms, the performance and efficiency of a plant can be increased to match the ever-changing world market.