Stokastik (R,s,S) ve Stokastik (R,S) Stok Kontrol Politikalarının Poliüretan Sektöründe Markov Karar Süreci Yardımıyla Karşılaştırılması

Bu çalışmada, olasılıklı stok politikaları tanıtılarak bunlardan periyodik gözden geçirmeye dayalı (R,s,S) stok politikasının Markov zinciri ve Markov karar süreci özelliği taşıdığı ispatlanmış, ardından bu politika; benzer özellikleri taşıyan (R,S) stok politikasıyla maliyetler açısından karşılaştırılmıştır. Türkiye'de poliüretan alanında faaliyet gösteren uluslar arası bir firmanın MDI (Metilendifeııil Diizosiyanat) hammaddesine ait verileri toplanmış ve bu stok kalemine ait talebin olasılıklı yapısı incelenmiştir. (R,s,S) stok politikasına uygun biçimde, durum uzayı tanımlanarak ve geçiş olasılıkları hesaplanarak Markov zinciri oluşturulmuştur. Bu politika, stok kaleminin maliyet fonksiyonu ile Markov karar problemi olarak gösterilmiş ve birim zamanda beklenen ortalama maliyet bu doğrultuda bulunmuştur. Ardından firmanın stok politikası olan (R,S) modeline aynı işlemler uygulanarak, maliyetler karşılaştırılmıştır.

Comparison of Stochastic (R,s,S) and Stochastic (R, S) Inventory Policies in Polyurethane Sector According to the Markov Decision Process

In this study, probabilistic inventory policies are introduced and it has been proved that the (R,s,S) inventory policy based on periodical revision shows Markov chain and Markov decision process characteristics. Thereafter, this policy has been compared with the (R,S) inventory policy which has similar features with the mentioned inventory policy, in terms of costs. The data of an international firm which operates in Turkey in the area of polyurethane, related to the raw material of MDI (methylenediphenyl diisosiyanathe) has been collected and the probabilistic structure of the demand related to this inventory item has been analyzed. The state space has been defined according to the (R,s,S) inventory policy, and the Markov chain has been formed by calculating the probabilities of transition. This policy has been shown as a problem of Markov decision, with the cost function of inventory item, and the expected avarage cost for a time unit has been found. Lastly, same processes have been applied to the (R,S) model which is the actual inventory policy of the mentioned firm, and the costs have been compared.

PDF

___

1.BARAY, Alp, Kasım 1993, "Bulanık Kümeler Kuramı ve İşletme Uygulamaları", I.U. İşletme Fakültesi Dergisi, Cilt: 22, Sayı: 2, s. 2.
2.BUCHAN, J., KOENIGSBERG, E., 1966, Scientific Inventory Management, New Delhi, Prentice-Hall of India Private Limited.
3.DAYAR, Tuğrul, 1994, "Stability and Conditioning Issues On The Numerical Solution Of Markov Chains", Doktora Tezi, North Carolina State University.
4.DOOB, J.L., 1990, Stochastic Processes, Canada,A Wiley-Interscience Publication, Wiley Classic Library Edition.
5.EHRHARDT, R., 1979, "The Power Approximation for Computing (s,S) Inventory PoliciesManagement Science, Volume: 30-5, s: 777-786.
6.EHRHARDT, R., 1984, "(s,S) Policies for a Dynamic Inventory Model with Stochastic Lead Times", Operations Research, Volume: 32-1, s: 121-132.
7.HILLIER, Frederick S., LIEBERMAN Gerald J.,2001, Introduction to Operation Research, United States of America, The McGraw-Hill Companies , Seventh Edition.
8.MEDHI, J., 2003, Stochastic Models in Queueing Theory, United States of America, Academic Press, Elsevier Science, Second Edition.
9.YIN, K. Karen, LIU, Hu ve JOHNSON, Neil E.,2002,"Marfcovian Inventory Policy with Application to the Paper Industry", Computers and Chemical Engineering, Volume: 26, s: 1399-1413.
10.PAPOULIS, A., 1984, Probability, Random Variables and Stochastic Processes, New York, McGraw Hill, Second Edition.
11.PUTERMAN, Martin L., 1994, Markov Decision Processes : Discrete Stochastic Dynamic Programming, United States of America, A Wiley Interscience Publication.
12.ROSS, Sheldon M., 2003, Introduction to Probability Models, United States of America, Academic Press, Elsevier Science, Eighth Edition.
13.SALDANA, Rafael P., CHANGHO, Marco Carlo C., "On Random Walk Models and Markov Chains", Proceedings of the Philippine Computing Science Congress, 2000, s. 204.
14.SCHNEIDER, H., 1978, "Methods for Determining the Re-Order Point of an (s,S) Ordering Policy When a Service Level is Specified", Journal of the Operation Research Society, Volume: 29(12), s: 1181-1193.
15.SILVER, Edward A, PYKE, David F. ve PETERSON, Rein, 1998, Inventory Management and Production Planning and Scheduling, United States of America, John Wiley & Sons, Third Edition.
16.T AHA, A. Hamdy, Ekim 2002, Yöneylem Araştırması, 6. Basımdan Çev. Ş. Alp Baray - Şakir Esnaf, İstanbul, Literatür Yayıncılık.
17.TAYLOR, Howard M., KARLIN Samuel, 1984,An Introduction To Stochastic Modeling, Orlando, Florida, Academic Press, Inc.
18.ÖZTÜRK, Ahmet, 2004, Yöneylem Araştırması,Bursa, Ekin Kitabevi.
19.WILSON, R.H., 1934, "A Scientific Routine for Stock Control", Harvard Business Review, 13.
20.WINSTON, Wayne L., 2004, Operation Research: Application and Algorithms, Canada, Thomson Brooks / Cole , Fourth Edition.
21.ZHENG, Yu-Sheng, FEDERGRUEN, A., Jul/Aug.1991, "Finding Optimal (s,S) Policies is About as Simple as Evaluating a Single Policy", Operations Research, Volume: 39-4, s: 654-665.