Derin öğrenme ile talep tahmini: Bir üçüncü parti lojistik firması için COVID-19 döneminde vaka analizi

COVID-19 pandemi döneminde yaşanan kısıtlamalar ve kapanmalar küresel tedarik zincirlerini büyük ölçüde etkilemiştir. Lojistik sektörü, bu süreçten en çok etkilenen sektörler arasında yer almaktadır. Bu nedenle, lojistik talebinin doğru ve hızlı tahmin edilmesi, etkin kaynak planlaması için önem taşımaktadır. Bu çalışmada, bir üçüncü parti lojistik firmasında COVID-19 pandemi dönemindeki talebin doğru tahmin edilmesi hedeflenmiştir. Lojistik firmasının Haziran 2020 ve Aralık 2020 tarihleri arasındaki sevkiyat verisi incelenmiştir ve tahmin problemi tek değişkenli zaman serisi olarak ele alınmıştır. Çalışma kapsamında, derin öğrenme tabanlı talep tahmini modeli önerilmiştir. Önerilen modelde evrişimli sinir ağı (CNN) ile uzun kısa dönem hafıza (LSTM) ağı bütünleştirilmiştir. CNN özniteliklerin çıkarılmasını, LSTM ağı ise uzun dönemli bağımlılıkların yakalanmasını sağlamaktadır ve önerilen model hibrit CNN-LSTM olarak adlandırılmıştır. Hibrit CNNLSTM’in tahmin performansı klasik tahmin yaklaşımlarının yanı sıra makine öğrenmesi ve derin öğrenme yaklaşımları ile karşılaştırılarak değerlendirilmiştir. Tüm tahmin yöntemlerinin parametre değerleri deneysel çalışmalar ile belirlenmiştir. Deneysel sonuçlara göre önerilen hibrit CNN-LSTM yöntemi diğer yöntemlerden daha yüksek tahmin performansı göstermiştir. Önerilen yaklaşım, lojistik talebinin doğru tahmin edilmesini sağlayarak işgücü ve kaynak planlaması faaliyetlerine girdi oluşturmaktadır.

Anahtar Kelimeler:

Talep tahmini, Üçüncü parti lojistik, Derin öğrenme, Evrişimli sinir ağları, Uzun kısa dönem hafıza

Demand forecasting with deep learning: Case study in a third-party logistics company for the COVID-19 period

The restrictions and closures experienced during the COVID-19 pandemic period have affected the global supply chains greatly. The logistics sector is among the most affected sectors from this process. For this reason, accurate and fast estimation of logistics demand is important for effective resource planning. In this study, the aim is to predict the demand accurately in a third-party logistics company during the COVID-19 pandemic period. The shipment data of a logistics company between June 2020 and December 2020 were examined, and the prediction problem was considered as univariate time series. In the scope of the study, a deep learning-based demand forecasting model is proposed. In the proposed prediction model, convolutional neural network (CNN) and long short-term memory (LSTM) network are integrated. CNN provides feature extraction, LSTM captures long-term dependencies, and the proposed model is called hybrid CNN-LSTM. The prediction performance of the hybrid CNN-LSTM was evaluated by comparing it with the classical prediction approaches as well as machine learning and deep learning approaches. The parameter values of all forecasting methods were determined by experimental studies. According to the experimental results, the proposed hybrid CNN-LSTM method showed higher performance than the other methods. The proposed approach generatesinput to workforce and resource planning activities by providing accurate estimation of logistics demand.

Keywords:

Demand forecasting, Third-party logistics, Deep learning, Convolutional neural networks, Long short-term memory,

PDF

___

[1] Waters D. Logistics An Introduction to Supply Chain Management. 1st ed. New York, USA, Palgrave Macmillan, 2003.
[2] Burnson P. “Top 50 Third-party logistics: Today’s marketplace not for the faint of heart”. Logistics Management, 59(6), 56-64, 2020.
[3] Council of Supply Chain Management Professionals. “CSCMP Supply Chain Management Definitions and Glossary”. https://cscmp.org/CSCMP/Educate/SCM_Definitions_an d_Glossary_of_Terms.aspx (15.12.2021).
[4] Barua L, Zou B, Zhou Y. “Machine learning for international freight transportation management: A comprehensive review”. Research in Transportation Business & Management, 34, 1-11, 2020.
[5] Yuan W, Chen JH, Cao JJ, Jin ZY. “Forecast of logistics demand based on grey deep neural network model”. 2018 International Conference on Machine Learning and Cybernetics, Chengdu, China, 15-18 July 2018.
[6] Gao Y, Chang D, Chen CH, Fang T. “Deep learning with long short-term memory recurrent neural network for daily container volumes of storage yard predictions in port”. 2018 International Conference on Cyberworlds, Singapore, 3-5 October 2018.
[7] Ren S, Choi TM, Lee KM, Lin L. “Intelligent service capacity allocation for cross-border E-commerce related thirdparty-forwarding logistics operations: A deep learning approach”. Transportation Research Part E: Logistics and Transportation Review, 134, 1-19, 2020.
[8] Yang C, Chang P. “Forecasting the demand for container throughput using a mixed-precision neural architecture based on CNN–LSTM”. Mathematics, 8, 1-17, 2020.
[9] Abosuliman SS, Almagrabi AO. “Computer vision assisted human computer interaction for logistics management using deep learning”. Computers and Electrical Engineering, 96, 1-12, 2021.
[10] Bousqaoui H, Slimani I, Achchab S. “Comparative analysis of short-term demand predicting models using ARIMA and deep learning”. International Journal of Electrical and Computer Engineering, 11(4), 3319-3328, 2021.
[11] Xia G, Ma Lu, Wang D, Sun Z. “Prediction of logistics demand based on grey neural network ensemble”. 2016 3rd International Conference on Systems and Informatics, Shanghai, China, 19-21 November 2016.
[12] Moscoso-Lopez JA, Turias IJ, Come MJ, Ruiz-Aguilar JJ, Cerban M. “Short-term forecasting of intermodal freight using ANNs and SVR: Case of the Port of Algeciras Bay”. Transportation Research Procedia, 18, 108-114, 2016.
[13] Yu N, Xu W, Yu KL. “Research on regional logistics demand forecast based on improved support vector machine: a case study of Qingdao City under the new free trade zone strategy”. IEEE Access, 8, 9551-9564, 2020.
[14] Yin Y, Chen S. “Research on the prediction model to the highway transportation demand based on moving average and grey theory”. International Conference on Grey Systems and Intelligent Services, Nanjing, China, 18-20 December 2007.
[15] Altın FG, Çelik Eroğlu Ş. “Gri tahmin ve Box-Jenkins yöntemleri ile Antalya Limani için aylık konteyner talep tahmini”. Mehmet Akif Ersoy Üniversitesi İktisadi ve İdari Birimler Fakültesi Dergisi, 7(3), 540-562, 2020.
[16] Fattah J, Ezzine L, Aman Z, Moussami HE, Lachhab A. “Forecasting of demand using ARIMA model”. International Journal of Engineering Business Management, 10, 1-9, 2018.
[17] Acı M, Ayyıldız Doğansoy G. “Demand forecasting for eretail sector using machine learning and deep learning methods”. Journal of the Faculty of Engineering and Architecture of Gazi University, 37(3), 1325-1339, 2021.
[18] Demir L, Akkaş S. “A comparison of sales forecasting methods for a feed company: A case study”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 24(4), 705-712, 2018.
[19] Kayapınar Kaya S, Yıldırım Ö. “A prediction model for automobile sales in Turkey using deep neural networks”. Journal of Industrial Engineering, 31(1), 57-74, 2020.
[20] Nacar EN, Erdebilli B. “Makine öğrenmesi algoritmaları ile satış tahmini”. Endüstri Mühendisliği Dergisi, 32(2), 307-330, 2021.
[21] Goodfellow I, Bengio Y, Courville A. Deep Learning. 1st ed. Cambridge, USA, MIT Press, 2016.
[22] Hochreiter S, Schmidhuber J. “Long short-term memory”. Neural Computation, 9(8), 1735-1780, 1997.
[23] Gers F, Schmidhuber J, Cummins F. “Learning to forget: continual prediction with LSTM”. Neural Computation, 12(10), 2451-2471, 2000.
[24] Karahan T, Nabiyev V. “Plant identification with convolutional neural networks and transfer learning”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 27(5), 638-645, 2021.
[25] Botchkarev A. “A new typology design of performance metrics to measure errors in machine learning regression algorithms”. Interdisciplinary Journal of Information, Knowledge and Management, 14, 45-79, 2019.
[26] Jierula A, Wang S, Oh TM, Wang P. “Study on accuracy metrics for evaluating the predictions of damage locations in deep piles using artificial neural networks with acoustic emission data”. Applied Sciences, 11(5), 1-20, 2021.
[27] Armstrong JS, Collopy F. “Error measures for generalizing about forecasting methods: Empirical comparisons”. International Journal of Forecasting, 8(1), 69-80, 1992.
[28] Martínez F, Frías MP, Pérez MD, Rivera AJ. “A methodology for applying k-nearest neighbor to time series forecasting”. Artificial Intelligence Review, 52(3), 2019-2037, 2019.
[29] Yi D, Ahn J, Ji S. “An effective optimization method for machine learning based on ADAM”. Applied Sciences, 10(3), 1-20, 2020.