Authors: Tin-Chih Toly Chen; Yu-Cheng Wang
Addresses: Department of Industrial Engineering and Systems Management, Feng Chia University, 100, Wenhwa Road, Seatwen, Taichung City 407, Taiwan ' Department of Aviation Mechanical Engineering, China University of Science and Technology, 200, Zhonghua St., Hengshan Township, Hsinchu County 312, Taiwan
Abstract: Cloud computing (CM) is bringing opportunities to various fields. In the manufacturing sector, cloud manufacturing (CMfg) borrows many useful concepts from CM. Among them, simulating a factory online is imperative. To this end, the time required for a simulation task needs to be estimated first. However, this topic has rarely been discussed. An artificial neural network (ANN) approach is proposed in this paper. In the proposed methodology, an ANN is constructed to estimate the required execution time for a simulation task. The real data of 90 simulation tasks have been collected to validate the proposed methodology. In addition, several existing methods were also applied to these tasks for a comparison. According to the experimental results, the proposed methodology outperformed the compared existing methods in improving the estimation accuracy. In addition, the planning horizon is the most decisive factor to estimating the execution time of a simulation task.
Keywords: cloud manufacturing; factory simulation; artificial neural networks; ANNs; workload estimation; cloud computing; task execution time; planning horizon; manufacturing industry.
International Journal of Internet Manufacturing and Services, 2014 Vol.3 No.4, pp.329 - 340
Received: 14 Mar 2015
Accepted: 29 Mar 2015
Published online: 20 Jul 2015 *