Optimizing machine utilization in semiconductor assembly industry using constraint-chromosome genetic algorithm

Semiconductor manufacturing is always aiming for an accurate capacity planning that is able to optimize the utilization of the resources especially in machine loading problems. There are two main approaches being studied to solve the problem: linear programming-based and bio-inspired approaches. Recently, more studies are focusing on bio-inspired approaches, where amongst them, genetic algorithm (GA) is being the most popular one. We propose a constraint-chromosome (CCGA) to solve this problem by applying the workable chromosome representation to the domain problem. The approach developed helps to avoid from getting trapped at local minima and is able to search for more solutions. This method is chosen to allow the running of GA in its original form as well as to ensure the computation is straightforward and simple. The objective of the algorithm is to optimize the utilization of the machines that leads to an increase in the company's overall profit. It has been found that the proposed CCGA is able to propose good solution to the problem.