Reinforcement Learning Combined With Radial Basis Function Neural Network to Solve Job-Shop Scheduling Problem
To complete jobs/tasks within their designated time periods, manufacturing companies utilize multiple machines. Job-shop scheduling is a critical element in job/task completion. This schedule consists of a sequence of doing consecutive jobs in a minimum amount of time. In addition, any conflict between the raw materials used in each job and its resource pool are to be avoided. This research applied the Reinforcement Learning (RL) method which is implemented in Temporal Difference Learning (TDL). Furthermore, the TDL focused on the Gradient-Descent method in which the Radial Basis Function Neural Network served as the approximation function. The input of this research was an initial critical path with no conflict-free schedule. Using the above methods, the conflict(s) could be eliminated gradually. Thus, the flexible job-shop scheduling can readily he made by any manufacturing company. Language used for this research is the Borland Delphi 7.0. All object structure and methods are made as easy as possible so that it can be implemented on the same problem with different application.
machine industry job-shop scheduling reinforcement learning radial basis function neural network
Ronald Suryaputra Williem Kuswara Setiawan
Faculty of Business Universitas Pelita Harapan Surabaya, CITO Superblock, Jl. Jend. A. Yani no.288, Faculty of Computer Science Hill University, Ormskirk, Lancashire, UK.Universitas Pelita Harapan Sur
国际会议
大连
英文
29-32
2011-07-10（万方平台首次上网日期，不代表论文的发表时间）