Title: Forecasting vehicular traffic in Philippine public transport networks: a neural controlled differential equation approach

Authors: Kyle Adrian A. del Castillo; Dharyll Prince M. Abellana; Miriam F. Bongo

Addresses: Department of Computer Science, College of Science, University of the Philippines Cebu, Gorordo Ave., Lahug, Cebu City, 6000, Cebu, Philippines ' Department of Computer Science, College of Science, University of the Philippines Cebu, Gorordo Ave., Lahug, Cebu City, 6000, Cebu, Philippines ' Industrial and Systems Engineering Department, Gokongwei College of Engineering, De La Salle University Manila, 2401 Taft Ave, Malate, Manila, 1004, Metro Manila, Philippines

Abstract: This study explores the application of neural controlled differential equations (NCDEs) for forecasting multivariate travel times within a road network in Metro Cebu (Philippines). The research uses a network comprising 20 routes in Cebu City, Philippines, with five key locations chosen as origin-destination pairs. The findings indicate that NCDEs significantly improve the accuracy of travel-time forecasting compared to traditional models. The promising results of this study suggest that hybrid models that combine NCDE with other methodologies could be developed to harness its robust performance, ensuring effectiveness when applied to irregular time-series data. Further validation and improvement of the applicability of these models in real-world traffic management systems could lead to more advanced AI-driven traffic management and control.

Keywords: travel time forecasting; neural controlled differential equations; NCDEs; neural networks; forecasting.

DOI: 10.1504/IJADS.2026.155065

International Journal of Applied Decision Sciences, 2026 Vol.19 No.4, pp.437 - 461

Received: 05 Mar 2025
Accepted: 08 Aug 2025
Published online: 25 Jul 2026 *

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