隨機克利金法在受限制下的模擬應用於週期性訂貨政策最佳化;Stochastic Kriging Metamodeling in Constrained Simulation Optimization :on Inventory Period Review Policy

NCUIR > School of Management at National Central University > Graduate Institute of Industrial Management > Electronic Thesis & Dissertation > Item 987654321/76542

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/76542

Title:	隨機克利金法在受限制下的模擬應用於週期性訂貨政策最佳化;Stochastic Kriging Metamodeling in Constrained Simulation Optimization :on Inventory Period Review Policy
Authors:	杜俊學;Du, Jun-Xue
Contributors:	工業管理研究所
Keywords:	蒙地卡羅模擬;克利金內插法;隨機克利金法;(s , S )存貨系統
Date:	2018-07-20
Issue Date:	2018-08-31 11:27:03 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨機克利金法(Stochastic Kriging)是一種開發已久，用於模型化(metamodeling)隨機模擬模型的方法。隨機克利金法是將克利金法(Kriging)進行改良後的一種方法，相較於克利金法忽略了各模擬試驗中的每個重複試驗中輸出值的變異數，隨機克利金法加入了重複試驗中的內部變異數視為隨機模擬中固有的不確定性，進而使得隨機克利金法能夠更好的運用在隨機模擬上。本研究主要為研究隨機克利金法在受限制式下進行估計設計點的模擬值，建立一個能夠快速得出總成本的模型，在該模型下找出最佳的訂貨政策。實驗研究了在( s , S )存貨系統下，目標為找到s和S的最佳解，使其總成本最小化；使用隨機克利金法之後所得到的元模型(metamodel)，將我們給定的設計點帶入元模型中，篩選出服從限制式條件下的設計點，最後篩選出最小成本的s和S值。 ;This paper demonstrated an experiment exploring the potential of the Stochastic Kriging methodology for constrained simulation optimization. The experiment study an ( s , S ) inventory system with the objective of finding the optimal values of s&S and the minimize total cost. The goal function and constraints in these experiment, did the approach to determine the optimum combination predicted by the Stochastic Kriging prediction model. The results of this experiment indicate that Stochastic Kriging offers opportunities for solving constrained optimization problems in stochastic simulation
Appears in Collections:	[Graduate Institute of Industrial Management] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	199	View/Open

社群 sharing

Loading...