An ANFIS Based Vehicle Sales Forecasting Model Utilizing Feature Clustering and Genetic Algorithms

The automotive sector is one of Turkey’s most important industries, and the developments in technology are affecting the automotive sector as well as the other sectors. The methods that have been used to date indicate that the use of AI should be increased when the demand forecasting applications take into account the developments in the industry. For this purpose, by using the data taken from the Automotive Distributors Association and Turkish Statistical Institute Internet pages, intuitive learning hybrid ANFIS method is used to forecast the sales in this study. A clustering scheme is first applied to group the features, and then the features are fed into genetic algorithms to improve the prediction model performance. The experiments show that the prediction performance of the proposed method is good when compared to existing related studies in the literature.

ANFIS Tabanlı Öznitelik Kümeleme ve Genetik Algoritmaları Kullanan Bir Araç Satış Tahmini Modeli

Otomotiv sektörü Türkiye’nin en önemli sanayi kollarından biridir. Teknolojide yaşanan gelişmeler birçok sektörde olduğu gibi otomotiv sektörünü de etkilemektedir. Bugüne kadar yapılan çalışmalarda kullanılan yöntemler talep tahminindeki uygulamaların sanayide yaşanan gelişmeler dikkate alındığında yapay zekâ kullanımının artması gerektiğine işaret etmektedir. Bu amaçla Otomotiv Distribütörleri Derneği ve Türkiye İstatistik Kurumu İnternet sayfalarından alınan verileri kullanarak, bu çalışmada araç satışlarını tahmin etmek için sezgisel öğrenmeli melez ANFIS yöntemi kullanılmıştır. İlk once özellikleri gruplamak için kümeleme şeması uygulanır ve ardından tahmin modeli performansını geliştirmek için özellikler genetik algoritmalara beslenir. Deneyler, önerilen yöntemin tahmin performansının literatürdeki mevcut çalışmalarla karşılaştırıldığında iyi olduğunu göstermektedir.

PDF

___

[1] U. Kaya, A. Yılmaz, K. Keskin “Vehicle Sales Prediction Using Neural Fuzzy Logic Method in Industry 4.0,”, Industry 4.0 from the MIS Perspective, Editors: S. Gülseçen, Z. A. Reis, M. Gezer, Ç. Erol, Peter Lang Publication, vol 1, pp. 233-246, 2018.

[2] A. Khaled, R. Sanjay, S. Vineet, “An Efficient K-Means Clustering Algorithm” IPPS: 11th International Parallel Processing Symposium, 1998.

[3] T.T. Bilgin, Y. Çamurcu “Dbscan, optics ve K-means kümeleme algoritmalarının karşılaştırılması,” Politeknik Dergisi (Journal of Polytechnic), Vol. 8, No.2, 139-145, 2005.

[4] M. Pulat, İ. Deveci Kocakoç “Determining the Initial Population of Solving the Traveling Salesman Problem with Genetic Algorithms”, JOEEP: Journal of Emerging Economies and Policy, Vol. 2, No. 1, pp. 95-123, 2017.

[5] F.K. Wang, K.K. Chang, C.W. Tzeng “Using adaptive network-based fuzzy inference system to forecast automobile sales,” Expert Systems with Applications, Vol. 38, Issue 8, pp. 10587-10593, 2011.

[6] G. Tian, M. Zhou, J. Chu, B. Wang “Prediction models of the number of end-of-life vehicles in China” Proceedings of the 2013 International Conference on Advanced Mechatronic Systems, Luoyang, pp. 357-362, 2013.

[7] L. Yang, Li B. “The Combination Forecasting Model of Auto Sales Based on Seasonal Index and RBF Neural Network” International Journal of Database Theory and Application, Vol. 9, pp. 67-76, 2016.

[8] A. Vahabi, S.S. Hosseininia, M. Alborzi “A Sales Forecasting Model in Automotive Industry using Adaptive Neuro-Fuzzy Inference System(ANFIS) and GA(GA)” (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 7, No. 11, pp. 24- 30, 2016.

[9] D.S. Farahani, M. Momeni, N.S. Amiri “Car Sales Forecasting Using ANNs and Analytical Hierarchy Process (Case Study Kia and Hyundai Incorporations in the USA)” Data Analytics 2016, The Fifth International Conference on Data Analytics, Venice, Italy, from October 9, 2016 to October 13, pp. 57-62, 2016.

[10] U. Kaya, A. Yılmaz, K. Keskin “Vehicle Sales Prediction Using Neural Fuzzy Logic Method in Industry 4.0,” 4th International Management Information Systems Conference Industry 4.0 (IMISC2017), İstanbul University, İstanbul, Turkey, pp. 145-146, October 17-20, 2017.

[11] C.C. Lin, C.L. Lin, J.Z. Shyu “Hybrid multi-model forecasting system: A case study on display market” Knowledge-Based Systems, Vol. 71, pp. 279-289, 2014.

[12] M.Y. Chen “A hybrid ANFIS model for business failure prediction utilizing particle swarm optimization and subtractive clustering” Information Sciences, Vol. 220, pp. 180-195, 2013.

[13] L. Tan, S. Wang, K. Wang “A new adaptive network-based fuzzy inference system with adaptive adjustment rules for stock market volatility forecasting” Information Processing Letters, Vol. 127, pp. 32-36, 2017.

[14] E. Akkaya “ANFIS based prediction model for biomass heating value using proximate analysis components” Fuel, Vol. 180, pp. 687-693, 2016.

[15] B.S. Karthika, P.C. Deka “Prediction of air temperature by hybridized model (wavelet-ANFIS) using wavelet decomposed data” International conference on water resources, coastal and ocean engineering (ICWRCOE 2015), Aquatic Procedia, Vol. 4(2015), pp. 1155-1161, 2015.

[16] A. Sa-ngasoongsong, S.T.S. Bukkapatnam, J. Kim, P.S. Iyer, R.P. Suresh “Multi-step sales forecasting in automotive industry based on structural relationship identification” International Journal of Production Economics, Vol. 140, Issue 2, pp. 875-887, 2012.

[17] M.M. Ghiasi, A. Hajinezhad, H. Yousefi, A.H. Mohammadi “CO2 loading capacity of DEA aqueous solutions, Modeling and assessment of experimental data” International Journal of Greenhouse Gas Control, Vol. 56, pp. 289-301, 2017.

[18] G.S. Atsalakis, I.G. Atsalaki, C. Zopounidis “Forecasting the success of a new tourism service by a neuro-fuzzy technique” European Journal of Operational Research, Vol. 268, Issue 2, pp. 716-727, 2018.

[19] M.A. Soliman, H.M. Hasanien, H.Z. Azazi, E.E. El-kholy, S.A. Mahmoud “Hybrid ANFIS-GA-based control scheme for performance enhancement of a grid-connected wind generator”, IET Renewable Power Generation, Vol. 12, No. 7, pp. 832-843, 2018.

[20] J.T. Tsai, K.Y. Chiu, J.H. Chou “Optimal Design of SAW Gas Sensing Device by Using Improved Adaptive Neuro-Fuzzy Inference System” in IEEE Access, Vol. 3, pp. 420-429, 2015.

[21] J.T. Tsai, J.H. Chou, C.F. Lin “Designing Micro-Structure Parameters for Backlight Modules by Using Improved Adaptive Neuro-Fuzzy Inference System” in IEEE Access, Vol. 3, pp. 2626-2636, 2015.

[22] K. Kampouropoulos, F. Andrade, E. Sala, A.G. Espinosa, L. Romeral “Multi objective Optimization of Multi-Carrier Energy System Using a Combination of ANFIS and GAs” in IEEE Transactions on Smart Grid, Vol. 9, No. 3, pp. 2276-2283, May 2018.

[23] G. Tian, M. Zhou, J. Chu, B. Wang “Prediction models of the number of end-of-life vehicles in China” Proceedings of the 2013 International Conference on Advanced Mechatronic Systems, Luoyang, pp. 357-362, 2013.

[24] “How teh Genetic Algorithm Works”, mathworks.com, [Online]. Availabe: https://www.mathworks.com/help/gads/how-the-genetic-algorithm-works.html, [Accessed: June 27, 2019]

[25] A. Yücel “Tedarikçi Seçimi Probleminde Bütünleşik Sinirsel Bulanık Mantık Yaklaşımı” İstanbul Üniversitesi Endüstri Mühendisliği Bölümü, Doktora Tezi, İstanbul, 2010.

[26] Z. Şen “Mühendislikte Bulanık Mantık (Fuzzy) İle Modelleme Prensipleri” Su Vakfı Yayınları, İstanbul, 2004.

[27] R. Babuska “Fuzzy and Neural Control (Lecture Notes)” Delft, Delft University of Technology, 2001.

[28] O. Doğan “Uyarlamalı Sinirsel Bulanık Çıkarım Sisteminin (ANFIS) Talep Tahmini İçin Kullanımı ve Bir Uygulama” Doktora Tezi, Dokuz Eylül Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, Vol. 31, No. 1, pp. 257-288, 2016.

[29] “Virtual Library of Simulation Experiments: Test Funcions and Datasets”, sfu.ca, [Online]. Available: https://www.sfu.ca/~ssurjano/schwef.html, [Accessed: June 19, 2019].