Title: Modelling customer demand for mobile value-added services: non-stationary time series models or neural networks time series analysis?

Authors: Mohammad Hossein Vaghefzadeh; Behrooz Karimi; Abbas Ahmadi

Addresses: Industrial Engineering and Management Systems Faculty, Amirkabir University of Technology, Tehran, Iran ' Department of Industrial Engineering and Management Systems, Amirkabir University of Technology, Tehran, Iran ' Industrial Engineering and Management Systems Faculty, Amirkabir University of Technology, Tehran, Iran

Abstract: The present research applies two different modelling approaches to evaluate the historical demand for a special mobile value-added service (VAS) that is offered and delivered to airline customers. The first method is deterministic and includes non-stationary time series models that cover both mean and variance fluctuation, as well as seasonality effect, in the dataset. The second method is a metaheuristic approach in the form of artificial neural network time series analysis (ANN-TSA). These methods are used to evaluate the power of each category and to choose the best model based on appropriate criteria. The results show that non-stationary time series models outperform ANN-TSA, as indicated by the smaller number of errors in the simulation of the demand dataset.

Keywords: time series; analysis; non-stationary; artificial neural network; mobile value-added; seasonal effect; demand; forecasting.

DOI: 10.1504/IJISE.2023.129754

International Journal of Industrial and Systems Engineering, 2023 Vol.43 No.4, pp.555 - 581

Received: 25 Apr 2020
Accepted: 04 Apr 2021
Published online: 24 Mar 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article