An Improved Grouping Method for Multiple Fidelity Optimization in Simultaneous Scheduling of Machines and Vehicles

• Main Authors: Tsai, Yi Hsuan, 蔡宜璇
• Other Authors: James T. Lin
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/53889276555176732947

碩士 === 國立清華大學 === 工業工程與工程管理學系 === 104 === In different field of research there are many different fidelity models exist which considerd different system features. However, different fidelity modles have its pros and cons. When there is high fidelity model, we can evaluate system more accurately, but it will cause time-consuming and will lead to higher cost in building the model. On the other hand, lower fidelity model may exist bias, but it’s better for faster evaluation and the performance can provide partial of trend between low and high fidelity models. As the result, when facing multi-fidelity models, it is important to enhance the efficiency of optimization. In this research, we exploite Multi-fidelity Optimization with Ordinal Transformation and Optimal Sampling (MO2TOS) to solve the simultaneous scheduling problem of machines and automated guided vehicles (AGVs) in flexible manufacturing system (FMS). Based on the problem, we considering the zone-control and alternative machine features to setup the multi-fidelity model. In MO2TOS, grouping method is one of main factors which may affect the quality of optimization significantly. In this research, Improved Ordinal Transformation is proposed and for use with Global K-means to update the group after every iteration of MO2TOS. Futher more, we verify this method in different FMS layout and jobset. From the experimental results, this method can significantly allocate resource effectively and save simulation resources for higher and lower correlation models.