A Novel LSTM-Based Daily Airline Demand Forecasting Method Using Vertical and Horizontal Time Series

  • Boxiao Pan
  • Dongfeng YuanEmail author
  • Weiwei Sun
  • Cong Liang
  • Dongyang Li
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11154)


In this paper, we propose a LSTM-based model to cope with airlines’ needs for daily demand forecasting. For short-term (e.g. one day in advance) forecasting, we followed the traditional horizontal time series. But for long-term (e.g. half a month in advance) forecasting, the horizontal time series is no longer capable of doing this due to the lack of input data. So we came up with a novel vertical time series, which is also our main contribution in this paper. The vertical time series we propose possesses great application value and has big potential for future research. Empirical analysis showed that our LSTM-based model achieved the state-of-the-art prediction accuracy among all the tested models in both time series. Developed on a dataset from airline industry though, our approach can be applied to all sales scenarios where sale data is recorded continuously for a fixed period before the sale closes. So our research has big value for the industry.


  1. 1.
    Xie, G., Wang, S., Lai, K.K.: Short-term forecasting of air passenger by using hybrid seasonal decomposition and least squares support vector regression approaches. J. Air Transp. Manag. 37, 20–26 (2014)CrossRefGoogle Scholar
  2. 2.
    Wickham, R.R.: Evaluation of forecasting techniques for short-term demand of air transportation. Ph.D. thesis, Massachusetts Institute of Technology (1995)Google Scholar
  3. 3.
    Doganis, R.: Flying Off Course IV: Airline Economics and Marketing. Routledge, Abingdon (2009)Google Scholar
  4. 4.
    Van Ostaijen, T.: Dynamic booking forecasting for airline revenue management: a Kenya airways case study (2017)Google Scholar
  5. 5.
    Queenan, C.C., Ferguson, M., Higbie, J., Kapoor, R.: A comparison of unconstraining methods to improve revenue management systems. Prod. Oper. Manag. 16(6), 729–746 (2007)CrossRefGoogle Scholar
  6. 6.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  7. 7.
    Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 20(3), 273–297 (1995)zbMATHGoogle Scholar
  8. 8.
    Ho, T.K.: Random decision forests. In: Proceedings of the Third International Conference on Document Analysis and Recognition, vol. 1, pp. 278–282. IEEE (1995)Google Scholar
  9. 9.
    Quinlan, J.R.: Simplifying decision trees. Int. J. Man-Mach. Stud. 27(3), 221–234 (1987)CrossRefGoogle Scholar
  10. 10.
    Sutskever, I., Vinyals, O., Le, Q.V.: Sequence to sequence learning with neural networks. In: Advances in Neural Information Processing Systems, pp. 3104–3112 (2014)Google Scholar
  11. 11.
    Cho, K., et al.: Learning phrase representations using RNN encoder-decoder for statistical machine translation (2014). arXiv preprint arXiv:1406.1078

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Boxiao Pan
    • 2
  • Dongfeng Yuan
    • 1
    Email author
  • Weiwei Sun
    • 1
  • Cong Liang
    • 1
  • Dongyang Li
    • 1
  1. 1.Shandong UniversityJinanChina
  2. 2.South China University of TechnologyGuangzhouChina

Personalised recommendations