What Telco Data Can Tell Us About London's Tube Strike

Unacast Turbine is built to help Telcos process and enrich their data in order to create useful outputs and provide valuable insights around human mobility. This is done by taking anonymised user trace records, running it through our signal processing engine, and aggregating it to gain insight around population behavioural patterns.

What Telco Data Can Tell Us About London's Tube Strike

Read more about: Mature telcos are taking back the data throne from Big Tech...and building new revenue streams while they do it

Telco data is unique compared to other location data sources, because it generally provides a very high signal density. In other words, each individual device/cell phone supplies a high number of signals throughout the day. This is due to most devices being consistently active; making  phone calls, sending text messages, browsing the web, as well as just monitoring the network to find the strongest signal. As a result, telco data becomes a great source for detecting short term changes in behavioural patterns. When working with telcos that tend to have significant market share in their regions, it also facilitates realistic extrapolation and can thus be used to say something about an entire population.

This ability to understand human mobility by leveraging anonymised telco data has many interesting use cases. One terrific example was during the London tube strike that took place on June 6th 2022. The situation occurred when The National Union of Rail, Maritime and Transport Workers (RMT) called out 4,000 of their station staff members for a 24 hour strike on this date. Numerous stations were kept closed during the day, and Londoners had to find other means of transportation to get around the metropolis. It may seem obvious that this had a significant impact on the city’s overall infrastructure, but since the disruption was so spread out throughout the city, there are few ways of validating such an assumption. This is where location data, and more significantly; telco data, comes into play.

Figure 1. Anonymised user activity aggregated on h3 hexagons. The signals in this visualisation are all identified as part of a transit activity.

Thanks to Unacast Turbine’s collaboration with BT, we could draw valuable insights around the situation after processing their data through our platform. To give an idea of what the data looks like, Figure 1 is a visualisation where people’s activity has been aggregated on a h3 hexagon grid. It should tell something about the potential that the data holds in terms of human mobility and the transportation network.

It is then interesting to compare the day of the strike to a regular day. By doing so, it becomes clear that some parts of the tube network, such as The Piccadilly Line (Figure 2) saw a distinct decrease in activity, while parts of the road network saw a significantly higher load than usual (Figure 3)

Figure 2. Percentage difference of moving signals as compared to a regular day, aggregated on h3 hexagons around The Piccadilly Line. Red cells indicates an increased activity while blue cells indicates decreased activity.
Figure 3. Percentage difference of moving signals as compared to a regular day, aggregated on h3 hexagons around a segment of the M40/A40. Red cells indicates an increased activity while blue cells indicates decreased activity.

One analysis that then could be done, was to look at the overall increase in travel times during the tube strike. This clearly showed temporary disruption in people’s traveling behaviour, and below is a visualisation of the average percentage increase in travel time, from different Oyster Zones into Central London (Zone 1).

Figure 4. Percentage increase in travel time from different Oyster Zones into Central London, as compared to a regular day.

The largest difference was seen for journeys that started in Zone 3, where the average travel time increased by 30% compared to a regular day. Based on what can be observed in our historical data, the average travel time on a regular day is around 70 minutes for these journeys. This means that the average person spent approximately 21 minutes more to get into Central London on the day of the strike. The data also show that around 130,000 people made such a trip on the given day, which implies that the total time lost on journeys going from Zone 3 into Central London alone, can be estimated to 45,500 hours. Additionally, journeys from Zone 2 and Zone 4 increased by 23% and 21% respectively, and the total time lost for all zones combined, including journeys within Zone 1, can be estimated as high as 300,000 hours.

To put this in perspective, we can use an estimate derived from the Tag Data Book found at gov.uk. It suggests that the value of one hour travel time for commuters in the UK, can be estimated to about £10, taking inflation into account. This means that the financial loss caused by the increased travel times, can be estimated as high as £3 million.

A positive side effect that we could observe from the strike however, was that many people seem to choose walking as their alternative mode of transport. Below is a visualisation showing the increase in people walking around central London as compared to a regular day. We can see that there was a distinct increase, especially around Victoria, Pimlico and Vincent Square, where there was 37% more walking activity than usual.

Figure 5. Percentage difference of signals that has been identified as part of walking activity compared to a regular day, aggregated on h3 hexagons. Red cells indicated an increase while blue cells indicates a decrease.

Get the white paper

Schedule a Meeting

Meet with us and put Unacast’s data to the test.