top of page
mike59208

A pioneer in UK mobile network data for transport planning

Updated: Aug 31, 2021


Mobile Network Data used for transport planning and smart cities

Philippe Perret became a pioneer in using UK mobile network data to understand people’s movements through a twist of fate and the belief that there must be a better way of researching infrastructure projects. This is his story.


A twist of fate led Philippe Perret to become a pioneer in the application of mobile network data for urban and transport planning in the UK.


In 2000, after studying civil engineering in France, he followed his girlfriend (now wife) to Delft in the Netherlands, planning to do a Masters that would enable him to learn how to build tunnels.


Ten days before the course was due to start, however, the university told him that the course was cancelled and that he would have to select another one.


Choosing to go with transport and road engineering


Philippe chose transport and road engineering, the beginning of his career as a transport planner. Following his studies, he moved to the UK and, unable to find work in tunnelling, joined transport consultancy Scott Wilson, where he developed transport models to assess new road and rail schemes, national infrastructure masterplans for countries such as Montenegro.


Philippe quickly realised that the underlying transport data he was using for such schemes was inadequate. "Sometimes we were using roadside interviews of 500 people to extrapolate movement across the UK or any other country, which was a bit of a nonsense," he said.


Mobile Network Data - the solution


While working on a project relating to the M6 motorway, he used alternative approaches of consisting of GPS traces and mobile network data (MND) to develop a local model to test a segment of the project alongside traditional methods such as roadside interviewing, which involves stopping drivers and questioning them about their journeys.


"The answers from the various models were different, but not totally different," he said. "They wouldn't have changed the outcome of the scheme. However, it was obvious that we could make better decisions using better data sets by better estimating the key metrics linked to proposed schemes."


"I thought there was really something there. The data would give us powerful insights because the volume of observation data was so high"

An encounter with Citi Logik


In another twist of fate, he met two of the founders of Citi Logik, John Rands and Nick Bromley, as they had just founded the business using data from Vodafone. "I thought there was really something there. The data would give us powerful insights because the volume of observation data was so high" he said but he realised that to make it working it required inputs from transport experts.


Citing a project he worked on using traditional methods, Philippe recalled that he once used 26,000 actual trips to extrapolate 66m trips - so that 0.04 per cent of the movements were observed and 99.96 per cent were synthesised. "I thought why are we bothering?"


By contrast, using mobile phone data, 10m observations can be taken in two weeks, whereas GPS can record 1m observations in a year and roadside interviewing can take only a statistically insignificant sample.


Saving money for transport planners


In 2015, Philippe joined Citi Logik. One of his motivations was to enhance the status of transport planners. "There was a perception that transport planners were expensive and used guesswork in their models that are used to make decisions that cost millions in taxpayer money. If we could make the data more robust, we could make planners better respected."


"Instead of being seen as people playing with data, our customers saw that we were trying to understand the implications from a transport perspective, that we understood the language transport planners used and the problems they had."

His transport background helped to establish Citi Logik's credibility in the sector. "Instead of being seen as people playing with data, our customers saw that we were trying to understand the implications from a transport perspective, that we understood the language transport planners used and the problems they had."


The potential of Mobile Network Data


He believes that mobile network data has a huge amount of untapped potential. "We are still in the infancy of getting MND understood. It is clear that the data is offering far more insights than we had before. Working with mobile operators, we are sampling one in four people in the UK. Covid has made it obvious that this kind of data can be used to understand how people move. It allows you to segment numbers in many more ways than surveys so you can understand more about changing behaviour."


For example, unlike roadside interviews, MND shows whether people are static, which helps planners understand how many people are working from or near home. This helps them forecast demand for rail journeys, which is likely to be impacted by concerns over social distancing.


Philippe also sees potential in the future to use other forms of data. "At the moment we are focused on mobile phone data, but we are also able to combine it with other data sets. It is important that we keep an open mind and embrace emerging technologies, which may offer insights as powerful as those from MND or enhancing insights from MND using data fusion."

161 views0 comments

Recent Posts

See All

Comments


bottom of page