
SOM II, sub-project: Travel habits - data integration
In autumn 2020, volunteer Lunds could participate in a new survey on travel habits. It was Trivector and Future by Lund who again collected data through the TravelVu app. There were several purposes for the investigation. One was to understand how to travel in the new district of Brunnshög, one was to follow up on the municipality's objectives, one was to be able to design the transport system to meet the needs of travellers. Finally, they also wanted to compare how the data from the TravelVu app is consistent with the traditional point measurement.
The development of the new district of Brunnshög is progressing rapidly, and the municipality has a clear target that the proportion of car journeys to/from the area should be at most one third of all journeys. (Brunnshög's sustainability goals). In order to reach the goal, it becomes important both to follow up towards the goal, but also to understand how the municipality can meet the needs of travellers. The aim was to combine data collection among residents and workers at Brunnshög with analyses of how well the routes TravelVu collects match the municipality's calculations of traffic. By combining new ways of collecting data (TravelVu) with the municipality's traditional measurements (point counts), a better overall picture of traffic flows can be created.
When COVID-19 struck, many of the people working at Brunnshög started working from home. It was simply not possible to conduct travel survey among employees at Brunnshög. Instead of focusing on Brunnshög, it was decided to collect data throughout Lund.
Travel habits were examined using the TravelVu app downloaded in a regular smartphone. The data collection took place on 14/9-1/11 2020 and the participants collected data on how they travel for a week or longer. The data collection was both part of traffic safety research funded by the Swedish Transport Agency's Signage Fund and in the SOM project.
One hope was to gain knowledge that would help shape the transport of the future. The idea was to be able to compare two cycling surveys with each other, and to see differences between what it looked like before the pandemic and in the middle of the pandemic. In addition, there was curiosity about how the construction of the tramway would have affected travel. The participation in this survey was unfortunately low (just under a hundred participants), probably due to COVID-19, and the data was deemed insufficient to conduct this type of survey.
The second part of the sub-project, combining TravelVU data with traditional data, could still be implemented. The starting point was data collected in the first sub-project on bicycle journeys in SOM. These could be compared with the municipality's bicycle bills from the same period with very good results. The compliance was surprisingly large.
With traditional bicycle counts, the municipality counts how many people pass a point. This means that the municipality knows a lot about what it looks like in that point, but not what cycling looks like between the different points.
With TravelVU data, we know how an individual moves in the traffic system, but only a small proportion of all road users are involved in collecting data. By combining new and old ways of collecting data, a better overall picture can be created.
What was the result
In the project, Trivector worked together with both Lund Municipality, which contributed data from traffic counts, and with LTH, which carried out the analyses. The results showed that the cycling data collected by TravelVu is highly consistent with the bicycle volumes counted by the municipality at different intersections during the same time.
The analysis compared the number of cycle passes from TravelVU data with the number counted in the municipal censuses. Analyses were performed both for all accesses to the intersections (just over 450 data points) and with pooled values, one per intersection (just over 100 points). The results show that TravelVu can explain between 80 and 90 percent of the variation in what was counted manually. This means that there is a good agreement how data from TravelVu shows that you cycle in Lund and how you have actually cycled during the same time.
“These are absolutely amazing results,” said Emeli Adell after the project, we actually didn't think it would be so good. This means that for the first time we can actually draw a good picture of what the cycle flows in Lund look like. This gives us the opportunity to continue to work on understanding how to cycle in Lund and to create good conditions for cyclists. We have now also received additional projects funded by Vinnova where we will continue to investigate just this. The project is called “Finding the untapped cycle potential” and is part of Lund Open Sensing City. We are also curious if compliance is equally good for pedestrians, which we hope to investigate in future projects as well. If it turns out that TravelVu can collect geographic data that is geographically representative of how people actually move in the city, it would be a breakthrough and provide great opportunities for both municipalities and researchers.
Facts Travel habits - data integration
The project is a sub-project of the SOM project is part of the strategic innovation programme for the Internet of Things, IoT Sweden, which is funded by Vinnova. Anders Trana at Future by Lund was the project manager for the entire project. The project started on 1 September 2017 and ran until December 2020.
Project Time: 2020-02-01 - 2020-12-18
Project Manager: Emeli Adell, Trivector
Partnerschappen: Trivector, Lund University, Lund Municipality
