Prediction of delivery and pickup times and routes in urban areas using spatio-temporal neural networks and basic reinforcement learning

Alejo Cardenas, Hugo Francisco (2025) Prediction of delivery and pickup times and routes in urban areas using spatio-temporal neural networks and basic reinforcement learning. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (541kB) | Preview

Abstract

This thesis investigates how well spatio-temporal neural networks may help to better forecast pickup and delivery times and routes in cities and examines the use of basic reinforcement learning to maximize these choices. The study examines advanced models including Spatio-Temporal Graph Neural Networks (STGNN), hybrid LSTM with residual learning, and Long Short-Term Memory (LSTM) networks on actual metropolitan logistics data from five Chinese cities. With predictions confirmed via standard error measurements, a modular system was created to process, organize, and forecast trip times between back-to-back logistics nodes. Though not totally carried out inside the purview of this study, an offline reinforcement learning experiment was also conceived to choose routes according to these predictions. Results show that hybrid LSTM models do well in places where things aren't normal in cities, but STGNN models do better in cities where the roads are organised. This research provides insightful information for the best possible urban logistics and helps to expand the body of knowledge about how spatio-temporal neural networks and reinforcement learning can be used together in last-mile delivery systems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Ain, Qurrat Ul UNSPECIFIED
Subjects:	Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning H Social Sciences > HE Transportation and Communications > Urban Transportation
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	24 Aug 2026 14:40
Last Modified:	24 Aug 2026 14:40
URI:	https://norma.ncirl.ie/id/eprint/9615

Actions (login required)

View Item