The use of AI to unlock crowdsourced transport data

TSC Senior Technologist, Charles Carter, discusses an initiative using AI to unlock the benefits of crowdsourced transport data

I can still remember the reaction as I peered into the driver side window. Stone cold silence. I was conducting a roadside interview and had just asked the driver about the details of their current journey. The anger induced reticence was interrupted only by the intermittent beeps of frustrated drivers backing up into the distance. It was the late 2000s, and I had just started on my journey as a transport professional, collecting survey data to build transport models to help inform system improvements.

Fast forward to 2017, and this antiquated method of gathering journey information is still in use, albeit for building strategic models. Transport planners and modelers are calling out for a method that truly embraces the digital age. For smaller areas of study, short term installation of Automatic Number Plate Recognition (ANPR) cameras, supplemented with data from permanent Intelligent Transport Systems (ITS), is largely the method of choice. Traffic is not stopped, but, journey purpose and demographic information isn’t collected.

Mobile Phone Network Data (MND) and crowdsourced mobile phone GPS data are the two areas that provide most promise for a step change in journey insight information at scale. Transport professionals have explored the use of anonymised and aggregated MND to build strategic transport models in recent years, adding information from household surveys.

For crowdsourced GPS records, enter CATCH! (Citizens at the City’s Heart), an Innovate UK project involving TravelAi, TransportAPI, Elgin, Universities of Leeds (Consumer Data Research Centre) and Glasgow (Urban Big Data Centre), Transport Systems Catapult, The Behaviouralist, Oxfordshire County Council, Ipswich Borough Council, Coventry City Council, Newcastle City Council, and Leeds City Council.

The consortium is using a next generation ‘living’ journey planning app, called CATCH!, to crowdsource location based data. The latest AI algorithms, alongside sophisticated anonymisation techniques, provide detailed, multi-modal journey insights. Transport authorities, operators and other professionals can use these insights to plan and implement Intelligent Mobility in the UK, and further afield.

Getting a critical mass of users and therefore robust and reliable insights across the geographical area of interest to build models and make decisions from, is the big challenge. To help achieve this, the Local Authorities in the project are launching pilots to roll out the app for specific use cases, in specific areas, which will leverage the unique insight that CATCH! can bring. These range from the analyzing home to school travel and environmental impact, to residential travel plan monitoring.

Journey planning, dynamic journey management and data driven insights are all key building blocks for Mobility as a Service (MaaS), and CATCH! is a fledgling part of the ecosystem, ripe for acceleration.

The app can be downloaded from the App store or Google Play, and is currently in beta testing. It lets you plan journeys anywhere in the UK, and gives you info on live bus and train departures within a single click. Please download it if you are interested in providing feedback and taking part in this initiative for the benefit of the transport industry and public alike.

App store –

Google Play –

If you have any further questions, or want more info on the CATCH project, please email

Cookies on Catapult explained

To comply with EU directives we now provide detailed information about the cookies we use. To find out more about cookies on this site, what they do and how to remove them, see our information about cookies. Click OK to continue using this site.