Conference Papers 2016

Seminar
Day 2 (6 Oct 2016), Session 4, Big Data 1, 09:00 - 11:00

Status
Accepted, documents submitted

Submitted by / Abstract owner
Nicolae Duduta

Authors
Nicolae Duduta, Telefonica Digital, Nicholas Corby, Highways England, Stephen Rutherford, Jacobs

Short abstract
The paper provides an overview of the development of Highways England’s Trip Information System, a transport database covering the United Kingdom and built entirely using mobile phone data, in partnership with Telefonica / O2.

Abstract
In recent years, mobile phone cell ID data has emerged as an alternative to traditional data sources such as household travel surveys or roadside interviews (RSIs) for deriving demand data for transport models. Mobile phone data can provide up-to-date trip patterns for large samples of the population over any chosen geographic area, and the data can be collected automatically without inconveniencing the traveling public. While there is global research on the application of this new data source to transport modeling, this technology has been brought to the market primarily in the United Kingdom (UK).
After several proofs of concept in which mobile network operators such as Telefonica (trading as O2 in the UK) partnered directly with transport consultancies to build origin – destination (OD) matrices, the Trip Information System (TIS), commissioned by Highways England in 2015, has provided the first nationwide scale application of mobile phone data in transport planning.
In this paper, we provide an overview of the development of TIS, including some of the key aspects of the methodology, lessons learned, and the implications for transport modellers using the data.
We start by providing an overview of the data source – mobile phone cell events – and how they can be used to generate journey information. We review the existing literature on the topic, and we also draw from Telefonica’s previous experience in producing transport data from Cell ID information.
We then describe the overall objectives and specifications of TIS – a database covering the entire mainland UK and providing twelve months of journey data, by mode of transport, down to the level of Middle Layer Super Output Census Areas (MSOAs), along with capabilities such as select link analysis.
The paper then describes the methodology behind the TIS in more detail, with a focus on the main enhancements it provides compared to previous approaches to mobile phone data – including a reliability grading system for different users and journeys, and more advanced detection of modes of transport.
We conclude by discussing some of the key issues that transport modellers should be aware of when using TIS data, how to understand the strengths but also the limitations of the underlying technology, and how to make the best use of this innovative dataset.