Mund, Sumit K. (2016) Real-time analytics for urban road transport. Masters thesis, University of Huddersfield.

Urban road traffic congestion has been a constant problem both in UK and worldwide. Urban road transport authorities collect data from different sources. These data can be effectively utilised with an objective to minimize congestion and its impact. One of the ways for the same can be to find possible congestion in different routes beforehand and then plan accordingly either to reduce the effect or to avoid it entirely.

So this project aims to make effective use of existing data to predict journey time for near future e.g. 15/30/60 minutes ahead for different routes within the urban road traffic network. It also produced a working prototype for the journey time prediction with necessary data visualisations.

A complete data centric approach has been adopted to solve the problem of prediction by building a predictive model using machine learning algorithms with traffic volumes at different points as predictor and journey time for near future as the target. Given the nature and volume of the data, a big data platform (Apache Spark) was chosen as the analytics platform and the work also proposes a high level technical architecture for the end to end solution.

The results for journey time prediction for near future are quite encouraging with a consistency for different routes under the area of consideration.

__nas01_librhome_librsh3_Desktop_FINAL THESIS.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview


Downloads per month over past year

Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email