Deep reinforcement learning for optimal resource allocation

Abstract

With the increasing demand for goods in today’s world, manufacturers must find means to improve their productivity to meet these demands. Some ways to improve production are to use more advanced machinery or hire more manpower to meet the increasing demands. However, these methods can cause production costs to increase greatly which is unfavourable for manufacturers. Hence, there is a need to use methods that do not increase production cost to improve productivity such as optimizing scheduling of activities and resources in a production. In this project, a deep reinforcement learning scheduling algorithm will be developed by hybridizing current scheduling solutions to allocate resources of a manufacturing process and make it into a Graphic User Interface application for users to use the algorithm easily. This aims to provide users with an accessible and effective solution to their scheduling problems. In addition, this project will practice parameter tuning methods on the algorithm to obtain a parameter set that can achieve the most optimal result.