Havacilik Sektöründe Yedek Parçalarin Nicel Talep Tahmini: Karşilaştirmali Bir Analiz

Öz

İşletmeler, veriye dayalı planlar oluşturmalarını ve iyi bilgilendirilmiş iş kararları almalarını sağladığı için tahminlerden yararlanır. Arızalı ürün tahmini ise bir firma için gelecek arızalı parçaları tahmin ederek firmanın bu arızalara karşı hazırlıklı olmasını sağlar. Bu çalışmanın amacı; büyük bir havayolu bakım onarım firmasının en sık karşılaştığı 10 farklı ürün için ay bazında atölyelere gelen arızalı parçaların kantitatif tahmin teknikleri kullanılarak analiz edilmesi, en uygun tahmin modelinin belirlenmesi ve daha sonraki dönemlere ait arızalı ürün sayısını tahmin etmektir. Çalışmada, Ocak 2021 – Aralık 2022 dönemi verileri kullanılmıştır. Bu veriler ile 2023 yılı için arızalı ürün tahmini yapılmıştır. Dört farklı sayısal tahmin yöntemi kullanılmış ve bu dört farklı modelin tahmin etkinliği Ortalama Mutlak Hata (MAE), Ortalama Kare Hata (MSE) ve Ortalama Mutlak Yüzde Hata (MAPE) ölçüleri ile belirlenmiştir. Çalışma sonucunda; Basit Üstel Düzgünleştirme Yöntemi (α =0,90) en başarılı tahmin sonuçlarını verirken, Holt-Winters Yöntemi (α =0,50, β=0,30, γ=0,80) en başarısız tahmin sonuçlarını vermiştir. Sonuç olarak; bu çalışma arızalı parçaların tahmin edilmesinde kantitatif tahmin tekniklerinin kullanılmasının etkinliğini ortaya koymaktadır. Gelecekteki araştırmalarda tahmin doğruluğunu ve uyarlanabilirliğini daha da geliştirmek için farklı yöntemler veya makine öğrenimi algoritmaları çalışmaya dahil edilebilir.

Anahtar Kelimeler

• Talep Tahmini
• Hareketli Ortalamalar Yöntemi
• Basit Üstel Düzgünleştirme
• Holt Winters Yöntemi
• Holt’un Doğrusal Yöntemi.

Quantitative Demand Forecasting of Spare Parts in The Aviation Industry: A Comparative Analysis

Öz

Businesses benefit from forecasting, because it enables them to create data-driven plans and make well-informed business decisions. Defective product forecasting, on the other hand, predicts future defective parts for a company and ensures that the company is prepared for these shortcomings. This study's objective is to examine the defective parts arriving at the workshops monthly for 10 different products that are most frequently encountered by a large airline maintenance and repair company by using quantitative forecasting techniques, to determine the most appropriate forecasting model and to predict the number of defective products for the following periods. The study's data for the time frame January 2021 - December 2022 were used. With this data, defective product forecast for 2023 was made. Four different numerical forecasting methods were used and the forecasting efficiency of these four different models was determined by Mean Absolute Error (MAE), Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) measures. Because of the research, Simple Exponential Smoothing Method (α = 0.90) gave the most successful prediction results, while Holt-Winters Method (α = 0.50, β = 0.30, γ = 0.80) gave the least successful prediction results. Therefore, this study shows how effective it is to use quantitative prediction techniques in the prediction of defective parts. In future research, different methods perhaps the study can incorporate machine learning algorithms to advance improve prediction accuracy and adaptability.

Anahtar Kelimeler

• Demand Forecasting
• Moving Averages Method
• Simple Exponential Smoothing
• Holt Winters Method
• Holt's Linear Method.

Kaynakça

1. Aci, M., & Doğansoy, G. A. (2022). Demand forecasting for e-retail sector using machine learning and deep learning methods. Gazi University Journal of Engineering and Architecture, 37(3), 1325–1340. https://doi.org/10.17341/gazimmfd.944081
2. Aydın, M. Ç. (2017). Application of demand forecasting methods in clothing industry: A sample application (Master’s thesis, Selcuk University, Institute of Social Sciences).
3. Aydın, M. R. (2019). Demand forecasting with artificial neural networks: An application in retail sector. Istanbul Commerce University Journal of Science, 18(35), 43–55.
4. Bağcı, B. (2020). Grey system theory in forecasting prices of financial investment instruments. 3rd Sector Social Economy Journal. https://doi.org/10.15659/3.sektor-sosyal-ekonomi.20.03.1268
5. Bal, B. (2015). Demand forecasting and planning: Retail sector, e-commerce (Master’s thesis, Maltepe University, Institute of Social Sciences).
6. Bilişik, M. T. (2021). Comparison of artificial neural networks, regression, moving averages and Winters exponential smoothing methods in demand forecasting in the food industry. Euraslan Business & Economics Journal, 1–25.
7. Boylan, J., & Syntetos, A. (2006). Accuracy and accuracy implication metrics for intermittent demand. Foresight: The International Journal of Applied Forecasting, 4, 39–42.
8. Burçin, T. (2023). Analysis of vehicle loan demand forecast using artificial neural networks. Dumlupınar University Journal of Social Sciences, 78, 102–110. https://doi.org/10.51290/dpusbe.1298894

9. Can, M. (2009). Forecasting with time series analysis in businesses (Doctoral dissertation, Istanbul University).
10. Celebi, D., Bolat, B., & Bayraktar, D. (2009). Light rail passenger demand forecasting by artificial neural networks. In Proceedings of the 2009 International Conference on Computers & Industrial Engineering (pp. 239–243). https://doi.org/10.1109/ICCIE.2009.5223851
11. Çağlar, S. (2007). Principles to be complied with in interim financial statements, international financial reporting standards and Turkish application (Master’s thesis).
12. Çoban, F., & Demir, L. (2021). Demand forecasting with artificial neural networks and support vector regression: An application in food business. Dokuz Eylül University Journal of Engineering Sciences, 23(67), 327–338. https://doi.org/10.21205/deufmd.2021236729
13. Çuhadar, M. (2015). Modelling of foreign tourism demand for Muğla province and forecast for 2012–2013. International Journal of Economic and Administrative Studies, 12. https://doi.org/10.18092/ijeas.96623
14. Demirci, N. (2015). Application and evaluation of demand forecasting methods in the glass industry (Master’s thesis, Istanbul Technical University).
15. Dinçer, E., & Ekin, E. (2017). Solution approach to spare parts inventory management problems in aviation industry with linear programming. Journal of Life Economics, 4(2), 77–102.
16. Doğan, O. (2016). The use of adaptive neuro fuzzy inference system (ANFIS) for demand forecasting and an application. Dokuz Eylül University Journal of Economics and Administrative Sciences, 31(1), 257–288. https://doi.org/10.24988/deuiibf.2016311513
17. Ertuğrul, İ., & Özçil, A. (2017). Comparative analysis of individual consumer loan demand forecasts with vector auto regression and artificial neural network models. Journal of Financial Research and Studies, 9(16), 19–38. https://doi.org/10.14784/marufacd.305559
18. Es, H., Kalender, F. Y., & Hamzaçebi, C. (2014). Türkiye net energy demand estimation with artificial neural networks. Gazi University Journal of Engineering and Architecture, 29(3). https://doi.org/10.17341/gummfd.41725
19. Ferreira, R. P., Martiniano, A., Ferreira, A., Ferreira, A., & Sassi, R. J. (2016). Study on daily demand forecasting orders using artificial neural network. IEEE Latin America Transactions, 14(3), 1519–1525. https://doi.org/10.1109/TLA.2016.7459644
20. İnallı, K., Işık, E., & Dağtekin, İ. (2014). Estimation of efficiency and production parameters in Karakaya HEPP with artificial neural network. Dicle University Journal of Engineering, 5(1), 59–68.
21. Kandananond, K. (2011). Forecasting electricity demand in Thailand with an artificial neural network approach. Energies, 4(8), 1246–1257. https://doi.org/10.3390/en4081246
22. Karaoğlanlı, A. C. (2021). Maintenance, repair and renewal activities in civil aviation, aviation 4.0 and new trends. Manufacturing Technologies and Applications, 2(1), 61–74.
23. Kılıç, F., Akkaya, M. R., & Memili, N. (2018). Daily demand forecast for cafeteria using artificial intelligence techniques. European Journal of Science and Technology, 13, 65–71. https://doi.org/10.31590/ejosat.397549
24. Kılıç, S. S., & Sözen, A. (2023). The use of moving average indicator used in forex markets as an expert advisor in transactions with energy and mining products. Polytechnic Journal. https://doi.org/10.2339/politeknik.544135
25. Makridakis, S., & Wheelwright, S. C. (1977). Forecasting: Issues and challenges for marketing management. Journal of Marketing, 41(4), 24–38. https://doi.org/10.2307/1250231
26. Majed Saleh Al-Hafid, H., & Al-Maamary, G. H. (2012). Short term electrical load forecasting using Holt–Winters method. AL-Rafdain Engineering Journal, 20(6), 15–22. https://doi.org/10.33899/rengj.2012.63377
27. Özdemir, A., & Özdemir, A. (2006). Comparison of methods used in demand forecasting: Ceramic product group company application. Ege Academic Review, 6(2), 105–114.
28. Özger, Y. E., Akpinar, M., Musayev, Z., & Yaz, M. (2019). Estimation of electric charge by genetic algorithm-based Holt–Winters exponential smoothing method. Sakarya University Journal of Computer and Information Sciences, 2(2), 108–123. https://doi.org/10.35377/saucis.02.02.600620
29. Saatçioğlu, D., & Özçakar, N. (2016). Intermittent demand forecasting with artificial neural network method. Beykoz Academy Journal, 4(1), 1–32. https://doi.org/10.14514/BYK.m.21478082.2016.4/1.1-32
30. Sâri, T., Şensoy, S. R., Nurbaki, A. E., & Ağaç, İ. A. (2023). Demand forecasting with artificial neural networks approach: An application in metal goods manufacturing industry. Productivity Journal, 57(4), 701–718. https://doi.org/10.51551/verimlilik.1327524
31. Silver, E. A., Pyke, D. F., & Peterson, R. (2000). Inventory management and production planning and scheduling (3rd ed.). John Wiley & Sons.
32. Sönmez, O., & Zengin, K. (2019). Demand forecasting in food and beverage businesses: A comparison with artificial neural networks and regression methods. European Journal of Science and Technology, 302–308. https://doi.org/10.31590/ejosat.638104
33. Şahin, S., & Kocadağ, D. (2020). Literature research on demand forecasting in health sector. Düzce University Journal of Social Sciences, 10(1), 99–113.
34. Şahin, S., & Taşkesen, Ş. (2022). Demand forecasting studies in transportation and tourism sector: A literature review. International Journal of Business Science and Applications, 2(2), 147–164.
35. Şimşek, S., & Uslu, H. (2023). Occupational health and safety practices and effects in aircraft maintenance-repair hangars. European Journal of Science and Technology, 50, 178–189.
36. Terzioğlu, E., & Şahin, S. (2022). Demand forecasting studies in retail sector: Literature review. Düzce University Journal of Social Sciences, 12(2), 584–596. https://doi.org/10.55179/dusbed.1099085
37. Torun, Z., & Deste, M. (2021). An application to determine the appropriate demand forecasting method in material management in healthcare facilities. May 19 Social Sciences Journal, 2(3), 581–613. https://doi.org/10.52835/19maysbd.908786
38. Top, Y., & Yılmaz, E. (2009). Production management (2nd ed.). Yaprak Publishing.
39. Türk, E., & Kiani, F. (2019). Demand forecasting with artificial neural networks: White goods production planning example. Istanbul Sabahattin Zaim University Journal of the Institute of Science, 1(1), 30–37.
40. Türk Hava Yolları. (2011). 2011 activity report.
41. Ulucan, E., & Kızılırmak, İ. (2018). Demand forecasting methods in accommodation businesses: A research on artificial neural networks. Journal of Travel and Hotel Management, 15(1), 89–101. https://doi.org/10.24010/soid.415343
42. Yergök, D., & Acı, M. (2019). An alternative approach for daily demand forecasting in mass catering: Student regression. European Journal of Science and Technology, 64–73.