Belirsizlik altında tedarik zinciri ağ tasarımı: Çelik sektöründe hibrit kümeleme-optimizasyon yaklaşımı

Öz

Bu araştırma, Türkiye çelik sektöründe faaliyet gösteren bir işletmenin tedarik zinciri ağ tasarımı problemini tedarikçi güvenilirliği ve talep belirsizliği koşulları altında ele almaktadır. Üç aşamalı hibrit bir metodoloji önerilmiştir. İlk aşamada 12 tedarikçinin 12 aylık performans verileri Minitab 21 yazılımı kullanılarak K-ortalamalar kümeleme algoritması ile analiz edilmiş ve üç farklı performans grubu belirlenmiştir. İkinci aşamada, tedarikçi-fabrika-depo-müşteri yapısını içeren dört katmanlı tedarik zinciri ağı için LINGO 19.0 yazılımı ile karma tam sayılı doğrusal programlama modeli geliştirilmiş ve kümeleme sonuçları optimizasyon kısıtlarına entegre edilmiştir. Kümeleme kısıtının katkısını ölçmek amacıyla gerçekleştirilen baseline karşılaştırmasında, söz konusu kısıt olmaksızın düşük performanslı tedarikçi payının %52,4'e yükseldiği, kısıtın devreye girmesiyle %16,4'e gerilediği ve yalnızca %0,56 ek maliyet karşılığında 36,0 puanlık kalite kazanımı sağlandığı görülmüştür. Üçüncü aşamada, teslimat süreleri, talep seviyeleri ve kalite oranlarındaki belirsizlikler Python 3.11 programlama dili kullanılarak 10.000 yinelemeli Monte Carlo benzetimi ile modellenmiş ve risk metrikleri hesaplanmıştır. Bulgular, belirsizliğin deterministik çözüme göre %3,3 maliyet artışına neden olduğunu ve talep seviyesinin en kritik parametre olduğunu (elastikiyet: 1,185) göstermiştir. %95 güven düzeyinde Risk Değeri (VaR) 5.306.861 TL, Koşullu Risk Değeri (CVaR) 5.485.245 TL olarak hesaplanmıştır. Metodolojik katkı, istatistiksel kümeleme analizinin optimizasyon kısıtlarına parametrik entegrasyonunda ve bu entegrasyonun etkisinin nicel olarak ölçülmesinde yatmaktadır.

Anahtar Kelimeler

• Tedarik zinciri ağ tasarımı
• K-ortalamalar kümeleme
• karma tam sayılı programlama
• Monte Carlo benzetimi
• çelik sektörü

Supply chain network design under uncertainty: A hybrid clustering-optimization approach in steel industry

Öz

This study addresses the supply chain network design problem of a manufacturing company operating in the Turkish steel industry under supplier reliability and demand uncertainty conditions. A three-stage hybrid methodology is proposed. In the first stage, 12-month performance data of 12 suppliers are analyzed using the K-means clustering algorithm in Minitab 21 software, and three distinct performance groups are identified. In the second stage, a mixed-integer linear programming model is developed in LINGO 19.0 for the four-echelon supply chain network encompassing supplier-factory-warehouse-customer structure, and clustering results are integrated into optimization constraints. A baseline comparison conducted to quantify the contribution of clustering constraints reveals that removing these constraints raises the low-performance supplier share to 52.4%, whereas the constrained model reduces it to 16.4%, yielding a 36.0 percentage-point quality gain at only 0.56% additional cost. In the third stage, uncertainties in delivery times, demand levels, and quality rates are modeled through Monte Carlo simulation with 10,000 iterations using Python 3.11, and risk metrics are calculated. Findings reveal that uncertainty causes a 3.3% cost increase compared to the deterministic solution, and demand level is the most critical parameter (elasticity: 1.185). Value-at-Risk (VaR) at the 95% confidence level is calculated as 5,306,861 TL, and Conditional Value-at-Risk (CVaR) as 5,485,245 TL. The methodological contribution lies in the parametric integration of statistical clustering analysis into optimization constraints and the quantitative measurement of this integration's impact on network design decisions.

Anahtar Kelimeler

• Supply chain network design
• K-means clustering
• mixed-integer programming
• Monte Carlo simulation
• steel industry

Kaynakça

1. 1. Chen L., Dong T., Peng J. ve Ralescu D., Uncertainty analysis and optimization modeling with application to supply chain management: A systematic review, Mathematics, 11 (11), 2530, 2023.
2. 2. Jung J.Y., Blau G., Pekny J.F., Reklaitis G.V. ve Eversdyk D., A simulation based optimization approach to supply chain management under demand uncertainty, Computers & Chemical Engineering, 28 (10), 2087-2106, 2004.
3. 3. Paksoy T., Özceylan E. ve Weber G.-W., A multi-objective mixed integer programming model for multi echelon supply chain network design and optimization, Системні дослідження та інформаційні технології, (4), 43-55, 2010.
4. 4. Shankar B.L., Basavarajappa S., Chen J.C.H. ve Kadadevaramath R.S., Location and allocation decisions for multi-echelon supply chain network – A multi-objective evolutionary approach, Expert Systems with Applications, 40 (2), 551-562, 2013.
5. 5. You F., Grossmann I.E., Integrated multi-echelon supply chain design with inventories under uncertainty: MINLP models, computational strategies, AIChE Journal, 56 (2), 419-440, 2010.
6. 6. Benyoucef L., Xie X. ve Tanonkou G.A., Supply chain network design with unreliable suppliers: A Lagrangian relaxation-based approach, International Journal of Production Research, 51 (21), 6435-6454, 2013.
7. 7. Xiong G. ve Helo P., Challenges to the supply chain in the steel industry, International Journal of Logistics Economics and Globalisation, 1 (2), 160-175, 2008.
8. 8. Liu J., An R., Xiao R., Yang Y., Wang G., Wang Q., Implications from substance flow analysis, supply chain and supplier' risk evaluation in iron and steel industry in Mainland China, Resources Policy, 51, 272-282, 2017.

9. 9. Mohammadi M.A., Sayadi A.R., Khoshfarman M. ve Husseinzadeh Kashan A., A systems dynamics simulation model of a steel supply chain-case study, Resources Policy, 77, 102690, 2022.
10. 10. Askary M., Makui A. ve Tajik M., A robust multi-objective MIP model to optimize the supply chain of steel industry taking into account the sustainability approach, RAIRO-Operations Research, 58 (2), 1059-1091, 2024.
11. 11. Sabzevari Zadeh A., Sahraeian R. ve Homayouni S.M., A dynamic multi-commodity inventory and facility location problem in steel supply chain network design, The International Journal of Advanced Manufacturing Technology, 70 (9-12), 1267-1282, 2014.
12. 12. Antucheviciene J., Jafarnejad A., Amoozad Mahdiraji H., Razavi Hajiagha S.H. ve Kargar A., Robust multi-objective sustainable reverse supply chain planning: An application in the steel industry, Symmetry, 12 (4), 594, 2020.
13. 13. Dağdeviren M., Dönmez N., Kurt M., Developing a new model for supplier evaluation process for a company and its application, Journal of the Faculty of Engineering and Architecture of Gazi University, 21 (2), 247-255, 2006.
14. 14. Özel B., Özyörük B., Supplier selection with fuzzy axiomatic design, Journal of the Faculty of Engineering and Architecture of Gazi University, 22 (3), 415-423, 2007.
15. 15. Dağdeviren M., Erarslan E., Supplier selection using PROMETHEE sequencing method, Journal of the Faculty of Engineering and Architecture of Gazi University, 23 (1), 69-75, 2008.
16. 16. Öztürk M., Paksoy T., A combined DEMATEL-QFD-AT2 BAHP approach for green supplier selection, Journal of the Faculty of Engineering and Architecture of Gazi University, 35 (4), 2023-2044, 2020.
17. 17. Erik A., Kuvvetli Y., A new approach to supply chain performance assessment, Journal of the Faculty of Engineering and Architecture of Gazi University, 35 (4), 1923-1938, 2020.
18. 18. Ünlü A., Gezmişoğlu G., Çağıl G., Supplier evaluation with factor analysis based hybrid SWARA-VIKOR methods, Journal of the Faculty of Engineering and Architecture of Gazi University, 38 (4), 2231-2240, 2023.
19. 19. Shahabi M., Akbarinasaji S., Unnikrishnan A. ve James R., Integrated inventory control and facility location decisions in a multi-echelon supply chain network with hubs, Networks and Spatial Economics, 13 (4), 497-514, 2013.
20. 20. Bas S.A. ve Ozkok B.A., A fuzzy approach to multi-objective mixed-integer linear programming model for multi-echelon closed-loop supply chain with multi-product multi-time period, Operations Research and Decisions, 30 (1), 25-46, 2020.
21. 21. Kabadurmus O., Erdogan M.S., Sustainable, multimodal and reliable supply chain design, Annals of Operations Research, 292 (1), 47-70, 2020.
22. 22. Mishra S. ve Shaw K., Integration of capacitated lot sizing decision and supply chain network design model addressing sustainable and risk factors, Journal of Modelling in Management, https://doi.org/10.1108/JM2-11-2024-0370, 2025.
23. 23. Zegai R., Khettabi I. ve Benyoucef L., Multi-criteria resilient and sustainable supply chain network redesign under uncertainty: A hybrid fuzzy bounded confidence approach, Process Integration and Optimization for Sustainability, https://doi.org/10.1007/s41660-025-00413-6, 2025.
24. 24. Day M., Magnan G.M. ve Moeller M.M., Evaluating the bases of supplier segmentation: A review and taxonomy, Industrial Marketing Management, 39 (4), 625-639, 2010.
25. 25. Rezaei J. ve Wang J., Linking supplier development to supplier segmentation using best worst method, Expert Systems with Applications, 42 (23), 9152-9164, 2015.
26. 26. Azadnia A.H., Ghadimi P., Saman M.Z.M., Wong K.Y. ve Sharif S., Supplier selection: A hybrid approach using ELECTRE and fuzzy clustering, Informatics Engineering and Information Science (Communications in Computer and Information Science), 252, 663-676, 2011.
27. 27. Er Kara M. ve Oktay Fırat S.Ü., Supplier risk assessment based on best-worst method and k-means clustering: A case study, Sustainability, 10 (4), 1066, 2018.
28. 28. Izadikhah M., Saen R.F., Ahmadi K. ve Shamsi M., How to use fuzzy screening system and data envelopment analysis for clustering sustainable suppliers? A case study in Iran, Journal of Enterprise Information Management, 34 (1), 199-229, 2021.
29. 29. Anari E.T. ve Zegordi S.H., The type of supplier involvement in new product development in the automotive industry: Metaheuristic-based K-means clustering and analytic hierarchical process methods, Journal of Advances in Management Research, 22 (1), 90-110, 2024.
30. 30. Husna A., Ghasempoor A. ve Amin S.H., A proposed framework for supplier selection and order allocation using machine learning clustering and optimization techniques, Journal of Data, Information and Management, 6 (3), 235-254, 2024.
31. 31. Nasiri G.R., Miandoabchi E. ve Javadi M., Developing a two-stage decision-making method for selecting and clustering suppliers based on the resilience criteria, Group Decision and Negotiation, 34 (1), 7-34, 2024.
32. 32. Saugo R.A., Lima Junior F.R., Carpinetti L.C.R., Duarte A.P. ve Peinado J., An approach for sustainable supplier segmentation using adaptive network-based fuzzy inference systems, Mathematics, 13 (19), 3058, 2025.
33. 33. Kim J., Lee Y. ve Moon I., Optimization of a hydrogen supply chain under demand uncertainty, International Journal of Hydrogen Energy, 33 (18), 4715-4729, 2008.
34. 34. Almansoori A. ve Shah N., Design and operation of a stochastic hydrogen supply chain network under demand uncertainty, International Journal of Hydrogen Energy, 37 (5), 3965-3977, 2012.
35. 35. Dai Z. ve Zheng X., Design of close-loop supply chain network under uncertainty using hybrid genetic algorithm: A fuzzy and chance-constrained programming model, Computers & Industrial Engineering, 88, 444-457, 2015.
36. 36. Shabani N. ve Sowlati T., Evaluating the impact of uncertainty and variability on the value chain optimization of a forest biomass power plant using Monte Carlo simulation, International Journal of Green Energy, 13 (7), 631-641, 2016.
37. 37. Salem R.W. ve Haouari M., A simulation-optimisation approach for supply chain network design under supply and demand uncertainties, International Journal of Production Research, 55 (7), 1845-1861, 2017.
38. 38. Rohaninejad M., Sahraeian R. ve Tavakkoli-Moghaddam R., Multi-echelon supply chain design considering unreliable facilities with facility hardening possibility, Applied Mathematical Modelling, 62, 321-337, 2018.
39. 39. Ozkan O. ve Kilic S., A Monte Carlo simulation for reliability estimation of logistics and supply chain networks, IFAC-PapersOnLine, 52 (13), 2080-2085, 2019.
40. 40. Bhosekar A., Badejo O. ve Ierapetritou M., Modular supply chain optimization considering demand uncertainty to manage risk, AIChE Journal, 67 (7), e17367, 2021.
41. 41. Thevenin S., Adulyasak Y. ve Cordeau J.-F., Material requirements planning under demand uncertainty using stochastic optimization, Production and Operations Management, 30 (2), 475-493, 2021.
42. 42. Amjath M., Eljack F. ve Haouari M., Stochastic optimization for strategic planning of efficient and sustainable hydrogen supply chain networks under demand uncertainty, Computers & Chemical Engineering, 202, 109298, 2025.
43. 43. Sabouhi F., Jabalameli M.S., Jabbarzadeh A. ve Fahimnia B., A hybrid approach for resilient sourcing and supply chain network design, Annals of Operations Research, https://doi.org/10.1007/s10479-025-06086-5, 2025.
44. 44. Zou S. ve Witthaut M., Assessment of supply chain designs considering uncertainty via a Monte Carlo simulation, Simulation in Produktion und Logistik 2025, 123-132, 2025.
45. 45. Jannatifar H., Network of supply chain in uncertainty condition and disturbances of suppliers (Case study: The steel supply chain), Specialty Journal of Engineering and Applied Science, 4 (1), 45-59, 2019.
46. 46. Pan X. ve Guo S., Dual-objective optimization of a green closed-loop supply chain in steel industry considering quantity discount, Annals of Operations Research, 355, 1989-2015, 2023.
47. 47. Kamalahmadi M. ve Mellat-Parast M., Developing a resilient supply chain through supplier flexibility and reliability assessment, International Journal of Production Research, 54 (1), 302-321, 2016.
48. 48. Yildiz H., Yoon J., Talluri S. ve Ho W., Reliable supply chain network design, Decision Sciences, 47 (4), 661-698, 2016.
49. 49. Benedito E., Martínez-Costa C. ve Rubio S., Introducing risk considerations into the supply chain network design, Processes, 8 (6), 743, 2020.
50. 50. Qazi A., Simsekler M.C.E. ve Formaneck S., Supply chain risk network value at risk assessment using Bayesian belief networks and Monte Carlo simulation, Annals of Operations Research, 322 (1), 241-272, 2022.
51. 51. Awan M.A.H., Naseem A., Ahsan A., Shahzad A., Ullah M. ve Sadiq A.S., Evaluating disruption scenarios for improving downstream oil supply chain resilience and cost minimization using Monte Carlo simulation, Scientific Reports, 15, 38910, 2025.
52. 52. Luo N. ve Tang C., Visual clustering of supply chain via collaborative annealing power K-means++ clustering, Proceedings of the 2024 4th International Conference on Computer Communication and Artificial Intelligence (CCAI), 689-693, 2024.
53. 53. Giannelos S., Konstantelos I. ve Strbac G., Optimal supply chain design using machine learning, risk assessment and optimisation applied to coal distribution, EURO Journal on Decision Processes, 13, 100062, 2025.
54. 54. Rahiminia M., Razmi J. ve Farahani S.S., Cluster-based supplier segmentation: A sustainable data-driven approach, Modern Supply Chain Research and Applications, 5 (3), 209-228, 2023.