請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/17954
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 楊曙榮(Shu-Jung Yang) | |
dc.contributor.author | Chia-Yu Yang | en |
dc.contributor.author | 楊家瑜 | zh_TW |
dc.date.accessioned | 2021-06-08T00:46:59Z | - |
dc.date.copyright | 2021-03-04 | |
dc.date.issued | 2021 | |
dc.date.submitted | 2021-02-18 | |
dc.identifier.citation | Adam N. Elmachtoub, Jason Cheuk Nam Liang Ryan McNellis (2020) Decision Trees for Decision-Making under the Predict-then-Optimize Framework, arXiv: 2003.00360. Boyd, S., Boyd, S. P., Vandenberghe, L. (2004). Convex Optimization. Cambridge university press. Bertsimas, D., Thiele, A. (2006). A robust optimization approach to inventory theory. Operations Research, 54(1), 150–168. Chen, D. S., Batson, R. G., Dang, Y (2011). Applied integer programming: modeling and solution. Cynthia Rudin Gah-Yi Ban (2018) The Big Data Newsvendor: Practical Insights from Machine Learning. Operations Research, 67(1) 99-108. Elmachtoub AN Grigas P (2017) Smart “predict, then optimize”. arXiv:1710.08005. Gallego, G., Ryan, J. K., Simchi-Levi, D. (2001). Minimax analysis for finite-horizon inventory models. IIE Transactions, 33(10), 861-874. Greff, K., Srivastava, R. K., Koutník, J., Steunebrink, B. R., Schmidhuber, J. (2016). LSTM: A search space odyssey. IEEE Transactions on Neural Networks and Learning Systems, 28(10), 2222-2232. Gao, F., Su, X. (2017). Manufacturing Service Operations Management. Manufacturing Service Operations Management, 19(1), 84-98. Jakob Hubera, Sebastian M¨ullerb , Moritz Fleischmannb Heiner Stuckenschmidt (2019) A data-driven newsvendor problem: From data to decision. European Journal of Operational Research, 278(3), 904-915. Jaynta Mandi, Emir Demirović, Peter. J Stuckey Tias (2020) Guns Smart Predict-and-Optimize for Hard Combinatorial Optimization Problems, AAAI Conference on Artificial Intelligence, 34(2) 1603-1610. Kwon, K., Cheong, T. (2014). A minimax distribution-free procedure for a newsvendor problem with free shipping. European Journal of Operational Research, 232(1), 234-240. Lawrence V Snyder Zuo-Jun Max Shen (2011) Fundamentals of Supply Chain Theory. John Wiley Sons. Matteo Sangiorgio Fabio Dercole (2020) Robustness of LSTM neural networks for multi-step forecasting of chaotic time series. Chaos, Solitons Fractals, 139 110045. Oroojlooyjadid, A., Snyder, L. V., Tak´ac, M. (2018) Applying deep learning to the newsvendor problem. IISE Transactions, 52(4) 444-463. Perakis, G., Roels, G. (2008). Regret in the newsvendor model with partial information. Operations Research, 56(1), 188-203. Parikh, N., Boyd, S. (2014). Proximal algorithms. Foundations and Trends in Optimization, 1(3), 127-239. Pantelis R. Vlachas, Wonmin Byeon, Zhong Y. Wan, Themistoklis P. Sapsis Petros Koumoutsakos (2018) Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long-Short Term Memory Networks. Proceedings of The Royal Society A: Mathematical, Physical and Engineering Sciences, 474(2213) 20170844-20170844. Punia, S., Singh, S. P., Madaan, J. K. (2020). From predictive to prescriptive analytics: A data-driven multi-item newsvendor model. Decision Support Systems, 136, 113340. Xiao Alison Chen Zizhuo Wang (2017) Optimal pricing for selling to a static multi-period newsvendor. Operations Research Letters, 45(5) 415-420. Zeilinger, M. N., Jones, C. N., Morari, M. (2011). Real-time suboptimal model predictive control using a combination of explicit MPC and online optimization. IEEE Transactions on Automatic Control, 56(7), 1524-1534. Zhang, Y., Gao, J. (2017). Assessing the performance of deep learning algorithms for newsvendor problem. In ICONIP 2017, 10634 LNCS, 912–92. Zhang, J. (2019). Gradient descent based optimization algorithms for deep learning models training. arXiv preprint arXiv:1903.03614. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/17954 | - |
dc.description.abstract | 採購與供貨成本控制一直是營運與供應鏈管理最重要的一環之一,許多領域廣泛地應用報僮模型與相關延伸來最小化成本,我們修改了經典報僮模型的基本假設,將原本為已知或是定值的商品價格設為隨機變數,利用相關變數來預測未來多期的價格,透過價格、需求與訂購數量之間的關係使成本最小化。大多數過去的研究會先利用統計或機器學習模型預測出所需要的參數,過程中並未考慮利用此參數推論決策後的損失;預測的最佳解未必是在做決策時的最佳解,因此,同步最佳化預測與決策的研究開始興起。在本論文,我們將報僮模型結合到此框架,並將其運用在深度學習的預測模型上,提供決策者一個較穩健並能最小化決策後成本的方法。 | zh_TW |
dc.description.abstract | Newsvendor modeling is widely applied in many industry contexts to manage supply under uncertainty. We amend the assumption of classical newsvendor models to operate under uncertain feature-based prices over time. We minimize inventory cost through exploiting the relationship among price, demand and order quantity. Most existing studies are conducted in the manner of first predicting the parameters and then using them to make better decision. However, the best solution for pred6iction may not be the best solution for the sake of decision-making. Therefore, the study of simultaneously optimizing both prediction and decision is getting the more attention recently. We integrate the newsvendor model into the predict-then-optimize framework though the loss function of deep learning. Our modeling framework is operated in the robust lens by minimizing the regret of making the decision upon the predictive parameters. | en |
dc.description.provenance | Made available in DSpace on 2021-06-08T00:46:59Z (GMT). No. of bitstreams: 1 U0001-0502202116574900.pdf: 4074661 bytes, checksum: 5ae12a70cff9297934868bc7171fdd58 (MD5) Previous issue date: 2021 | en |
dc.description.tableofcontents | 謝辭 i 摘要 ii Abstract iii Table of Contents iv List of Figures vii List of Tables viii 1.Introduction 1 1.1 Research Background 1 1.2 Research Purpose 2 2.Literature Review 4 3.Model Development 8 3.1 Problem description 8 3.2 Method components 9 3.2.1 Prediction 9 3.2.2 Decision 11 3.2.3 Use of mixed-integer linear programming for our model 14 3.3 Machine learning 16 3.3.1 Predict-then-optimize framework 16 3.3.2 Deep learning 16 4.Numerical Analysis 20 4.1 Dataset 20 4.2 Experimental design 20 4.2.1 The length of predict weeks 20 4.2.2 Exploration of cost indicators 21 4.2.3 Reference methods 23 4.2.4 Evaluation Criteria 26 4.3 Result 27 4.4 Synthetic data 33 4.4.1 Sensitivity to correlation between input and output data 33 4.4.2 Sensitivity to large outliers in price 37 5.Discussion 40 6.Conclusion 42 References 44 Appendix 47 Appendix A - Data 47 Raw material 47 Demand 115 Appendix B - Python code 118 Data preprocessing 118 Gurobi 120 Deep learning model 123 Comparing result 129 | |
dc.language.iso | en | |
dc.title | 同步最佳化預測和決策框架下的智造供貨管理 | zh_TW |
dc.title | Smart supply management under the predict-then-optimize framework | en |
dc.type | Thesis | |
dc.date.schoolyear | 109-1 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 謝凱宇(Kai-Yu Hsieh),陳立民(Li-ming Chen) | |
dc.subject.keyword | 存貨管理,智造營運管理,數據驅動最佳化,深度學習,機器學習, | zh_TW |
dc.subject.keyword | inventory management,smart operations,data-driven optimization,deep learning,machine learning, | en |
dc.relation.page | 133 | |
dc.identifier.doi | 10.6342/NTU202100606 | |
dc.rights.note | 未授權 | |
dc.date.accepted | 2021-02-18 | |
dc.contributor.author-college | 共同教育中心 | zh_TW |
dc.contributor.author-dept | 統計碩士學位學程 | zh_TW |
顯示於系所單位: | 統計碩士學位學程 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-0502202116574900.pdf 目前未授權公開取用 | 3.98 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。