• Capabilities: Business Advisory, Technology and Information, Digital Innovation

Using mobile phone data for transport planning

United Kingdom - Highways England’s Road Investment Strategy sets out a long-term approach for improving the country’s major roads and motorways. To help develop schemes identified as part of this improvement strategy, Highways England commissioned five regional traffic models to assess traffic flows and capacity on various route. These models covered the North, the Northern Powerhouse, the Midlands, the South West and South East of England. Appointed as Project Manager by Highways England, Arcadis was responsible for mapping the South West and South East regions, integrating the use of mobile phone data to better understand travel patterns across the area.

Mobile Data for Transport Planning

Each of the five regional models required an accurate ‘origin-destination’ (OD) trip matrix to be produced. 

Traditionally, these trip matrices have relied on data collected manually, including census surveys, road side interviews and household surveys. However, the upsurge in mobile phone ownership and mobile phone data offers a new and alternative data source. Compared to the conventional approach, the OD matrices developed with mobile phone data offer much wider geographical coverage, higher sample size, can take into account the day-to-day variability of trips and provide potential time and cost savings. 

However, due to the relative newness of this approach, no established practice or review of the technical issues around incorporating mobile phone data into OD trip matrices yet exists.

For example, privacy laws and other standards dictate that the OD data released by mobile phone companies must be pre-processed before it can be released. This process typically involves cleaning, anonymising, aggregating and then, to make it suitable for use in OD matrices, applying various algorithms. 

However, is this data then still fit for use?

This is just one of many issues that could impact on the quality of the data. However, working with a number of consortium partners, Arcadis developed a comprehensive approach to verifying and validating the pre-processed data to ensure that it could still be used and provide an accurate result. 

This included:

  • Bias detection and how to overcome it 
  • Missing data and how to overcome it
  • Validating the mobile phone dataset by comparing the information and insights with the data from conventional sources 
  • Vehicle modal detection
  • Reconciling the data collected from the urban environment as opposed to the rural areas

It was only once all of these requirements were satisfied that the data was incorporated into the OD matrices and used to produce the required traffic models.  

“The use of mobile phone data in transport planning offers much wider geographical coverage, higher sample size, can take into account the day-to-day variability of trips and provide potential time and cost savings over more conventional approaches.”

Dr Olga Feldman, Arcadis

Questions about this project

Olga Feldman

Technical Director - Transport Planning +44 (0)7 7256 82981 Ask me a question
Share on Wechat
"Scan QR Code" on WeChat and click ··· to share.