Abstract
A challenge for importers in the automobile industry is adjusting to rapidly changing market demands. In this work, we describe a practical study of car import planning based on the monthly car registrations in Austria. We model the task as a data driven forecasting problem and we implement four different prediction approaches. One utilizes a seasonal ARIMA model, while the other is based on LSTM-RNN and both compared to a linear and seasonal baselines. In our experiments, we evaluate the 33 different brands by predicting the number of registrations for the next month and for the year to come.
Funding statement: This work is supported by the Know-Center. The Know-Center is funded within the Austrian COMET Program – Competence Centers for Excellent Technologies – under the auspices of the Austrian Ministry of Transport, Innovation and Technology, the Austrian Ministry of Economics and Labor and by the State of Styria. COMET is managed by the Austrian Research Promotion Agency (FFG).
About the authors
Emanuel Lacic is a Software Engineer and Scientific Researcher in the Social Computing division at the Know-Center. He holds a M.Sc. and B.Sc. in Software Engineering and Information Systems from the University of Zagreb and, is currently a PhD student at the Graz University of Technology. His main research focus lies on Recommender Systems, Deep Learning and Social Network Analysis.
Matthias Traub studied Information and Computer Engineering at the Technical University of Graz. He was a research assistant of the department of Intelligent Information Systems at Joanneum Research (2013-2014). Since 2014 he is a member of the Social Computing department at the Know Center in Graz, working in the field of data driven systems with the focus on recommender systems.
Tomislav Duricic is a Software Engineer and a Junior Scientific Researcher in the Social Computing team at the Know-Center. He holds a Master’s degree in Software Engineering and Information Systems from the University of Zagreb and is currently a PhD student at the Graz University of Technology. His current research interests are in the fields of Complex Network Analysis and Recommender Systems.
Since October 2016 Eva Haslauer ist working as a Projectmanager at Porsche Austria GmbH & CO OG in Salzburg. She works in the Digital Department and is responsible for digital projects in the realm of connected car, used car sale and sales forecast. She holds a PhD in Applied Geoinformatics from the University of Salzburg and worked as PostDoc in Innovation Management at the GI Department.
Elisabeth Lex is assistant professor at Graz University of Technology and head of the Social Computing area at Know-Center. Her research focus is on Social Computing, Recommender Systems and Data Science and she has published 60+ papers in these fields. She earned her PhD in Text Mining in the Social Web at Graz University of Technology, Austria, advised by Michael Granitzer and Stefanie Lindstaedt.
Acknowledgment
The authors would like to thank Dieter Theiler and Dominik Kowald for their help and valuable comments on this work.
References
1. Jérôme Adda and Russell Cooper, The dynamics of car sales: A discrete choice approach, National Bureau of Economic Research, Report, 2000.10.3386/w7785Search in Google Scholar
2. James Bergstra and Yoshua Bengio, Random search for hyper-parameter optimization, Journal of Machine Learning Research 13(Feb) (2012), 281–305.Search in Google Scholar
3. George E. P. Box and Gwilym M. Jenkins, Time Series Analysis: Forecasting and Control, 1970.Search in Google Scholar
4. Peter J. Brockwell and Richard A. Davis, Introduction to time series and forecasting, Springer Science & Business Media, 2006.Search in Google Scholar
5. Rodney L. Carlson and M. Michael Umble, Statistical demand functions for automobiles and their use for forecasting in an energy crisis, Journal of Business (1980).10.1086/296080Search in Google Scholar
6. Paul Childerhouse, Stephen Michael Disney and Denis Royston Towill, On the impact of order volatility in the European automotive sector, Int. Journal of Production Economics (2008).10.1016/j.ijpe.2007.09.008Search in Google Scholar
7. Kyunghyun Cho, Bart van Merriënboer, Çağlar Gülçehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk and Yoshua Bengio, Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation, in: Proc. of EMNLP’14, ACL, October 2014.Search in Google Scholar
8. Volkan Ş. Ediger and Sertaç Akar, ARIMA forecasting of primary energy demand by fuel in Turkey, Energy Policy 35(3) (2007), 1701–1708.10.1016/j.enpol.2006.05.009Search in Google Scholar
9. Debra A. Elkins, Ningjian Huang and Jeffrey M. Alden, Agile manufacturing systems in the automotive industry, Int. Journal of Production Economics (2004).10.1016/j.ijpe.2003.07.006Search in Google Scholar
10. Florian Eyben, Felix Weninger, Stefano Squartini and Björn Schuller, Real-life voice activity detection with lstm recurrent neural networks and an application to hollywood movies, in: Proc. of ICASSP’13, IEEE, 2013.10.1109/ICASSP.2013.6637694Search in Google Scholar
11. Dean Fantazzini and Zhamal Toktamysova, Forecasting German car sales using Google data and multivariate models, Int. Journal of Production Economics (2015).10.1016/j.ijpe.2015.09.010Search in Google Scholar
12. Xavier Glorot and Yoshua Bengio, Understanding the difficulty of training deep feedforward neural networks, Aistats., 9 2010, 249–256.Search in Google Scholar
13. Alex Graves, Sequence transduction with recurrent neural networks, arXiv preprint arXiv:1211.3711 (2012).10.1007/978-3-642-24797-2Search in Google Scholar
14. Sepp Hochreiter and Jürgen Schmidhuber, Long short-term memory, Neural computation 9(8) (1997), 1735–1780.10.1162/neco.1997.9.8.1735Search in Google Scholar PubMed
15. Rob Hyndman and Yeasmin Khandakar, Automatic Time Series Forecasting: The forecast Package for R, Journal of Statistical Software 27(1) (2008), 1–22.10.18637/jss.v027.i03Search in Google Scholar
16. Rob J. Hyndman and Anne B. Koehler, Another look at measures of forecast accuracy, International Journal of Forecasting 22(4) (2006), 679–688.10.1016/j.ijforecast.2006.03.001Search in Google Scholar
17. Rafal Jorzefowicz, Wojciech Zaremba and Ilya Sutskever, An empirical exploration of recurrent network architectures.Search in Google Scholar
18. Xiaolei Ma, Zhimin Tao, Yinhai Wang, Haiyang Yu and Yunpeng Wang, Long short-term memory neural network for traffic speed prediction using remote microwave sensor data, Transportation Research Part C: Emerging Technologies 54 (2015), 187–197.10.1016/j.trc.2015.03.014Search in Google Scholar
19. Svante Mandell, Policies towards a more efficient car fleet, Energy Policy (2009).10.1016/j.enpol.2009.07.068Search in Google Scholar
20. Sean McNally, Predicting the price of Bitcoin using Machine Learning, 2016.Search in Google Scholar
21. Joe Miemczyk and Matthias Holweg, Building cars to customer order – what does it mean for inbound logistics operations?, Journal of Business Logistics (2004).10.1002/j.2158-1592.2004.tb00186.xSearch in Google Scholar
22. Tomáš Mikolov, Statistical language models based on neural networks, Presentation at Google, Mountain View, 2nd April (2012).Search in Google Scholar
23. Yurii Nesterov, A method of solving a convex programming problem with convergence rate O(1/k2), Soviet Mathematics Doklady 27(2) (1983), 372–376.Search in Google Scholar
24. Razvan Pascanu, Tomas Mikolov and Yoshua Bengio, On the difficulty of training recurrent neural networks, ICML (3) 28 (2013), 1310–1318.Search in Google Scholar
25. Koen Pauwels, Jorge Silva-Risso, Shuba Srinivasan and Dominique M. Hanssens, New products, sales promotions, and firm value: The case of the automobile industry, Journal of marketing 68(4) (2004), 142–156.10.1142/9789813229808_0008Search in Google Scholar
26. Anthony J. Robinson, An application of recurrent nets to phone probability estimation, IEEE transactions on Neural Networks 5(2) (1994), 298–305.10.1109/72.279192Search in Google Scholar PubMed
27. Peter Romilly, Haiyan Song and Xiaming Liu, Modelling and forecasting car ownership in Britain: a cointegration and general to specific approach, Journal of Transport Economics and Policy (1998), 165–185.Search in Google Scholar
28. E. Rudd, The Relationship between the National Income and Vehicle Registrations. Road Research Laboratory, RN 1518 (1951).Search in Google Scholar
29. Lisa Ryan, Susana Ferreira and Frank Convery, The impact of fiscal and other measures on new passenger car sales and CO 2 emissions intensity: evidence from Europe, Energy Economics 31(3) (2009), 365–374.10.1016/j.eneco.2008.11.011Search in Google Scholar
30. Haşim Sak, Andrew Senior and Françoise Beaufays, Long short-term memory based recurrent neural network architectures for large vocabulary speech recognition, arXiv preprint arXiv:1402.1128 (2014).10.21437/Interspeech.2014-80Search in Google Scholar
31. Raymond Y. C. Tse, An application of the ARIMA model to real-estate prices in Hong Kong, Journal of Property Finance 8(2) (1997), 152–163.10.1108/09588689710167843Search in Google Scholar
32. Mohammad Valipour, Mohammad Ebrahim Banihabib and Seyyed Mahmood Reza Behbahani, Comparison of the ARMA, ARIMA, and the autoregressive artificial neural network models in forecasting the monthly inflow of Dez dam reservoir, Journal of hydrology 476 (2013), 433–441.10.1016/j.jhydrol.2012.11.017Search in Google Scholar
33. Gerard Whelan, Modelling car ownership in Great Britain, Transportation Research Part A: Policy and Practice 41(3) (2007), 205–219.10.1016/j.tra.2006.09.013Search in Google Scholar
34. Peter Whitle, Hypothesis Testing in Time Series Analysis, Almqvist & Wiksells, 1951.Search in Google Scholar
35. Peter de Wolff, The demand for passenger cars in the United States, Econometrica: Journal of the Econometric Society (1938), 113–129.10.2307/1907143Search in Google Scholar
36. Wenzu Wu, Kunjin Chen, Ying Qiao and Zongxiang Lu, Probabilistic short-term wind power forecasting based on deep neural networks, in: Proc. of PMAPS’16, IEEE, 2016.Search in Google Scholar
37. Godfrey Yeung and Vincent Mok, Manufacturing and distribution strategies, distribution channels, and transaction costs: The case of parallel imported automobiles, Managerial and Decision Economics 34(1) (2013), 44–58.10.1002/mde.2573Search in Google Scholar
38. Wojciech Zaremba, Ilya Sutskever and Oriol Vinyals, Recurrent neural network regularization, arXiv preprint arXiv:1409.2329 (2014).Search in Google Scholar
© 2018 Walter de Gruyter GmbH, Berlin/Boston
