Digitalization is here, and with every passing second, enormous amounts of data are being created. Every other second, public transport vehicles send out their position. Large amounts of data can transform the way we work, but it requires that the data is easily accessible in order for us to create value from it. This is an example of how digitalization can contribute to significant savings through new knowledge and data-driven decisions 

When the solution is more expensive than the problem

The Swedish Transport Administration had in a previous study pointed out a road stretch as problematic due to significant variations in travel time between the bus stops, but they were uncertain about where the problem was. They planned to build a bus lane on each side of the road for approximately 8-9 million euros to solve the problem.  

To feel confident in the planned measure, it was decided to study the issue in more detail with the help of Flowmapper.

Data-driven analysis with Flowmapper

Flowmapper downloads GPS data from 90% of Swedish bus traffic every other second and has, over the years, built up Sweden’s largest public transport database. The data is quality-checked, linked together, and visualized in an analysis platform. The data is visualized in 25-meter-long sequences. 

By analyzing data from thousands of unique trips, it becomes clear where accessibility issues exist, how often they occur, and how much delay they cause. Accessibility problems are best identified by studying where variations in travel time occur. In the map below, the standard deviation for each 25-meter sequence is shown. When the sequences are red, it means that the travel time past that sequence varies greatly between different trips, i.e., there are accessibility problems on the stretch. When it’s green, the travel time past the sequence is consistent.  

From million-€ investment to smarter solution

The graph shows travel time variation hour by hour. From there, it is possible to see how fast the bus travels when it has free flow (10th percentile) and how much longer it takes for those who are most affected (90th percentile) during peak hours. In the example, the slowest buses during peak hours lose about 50 seconds compared to free flow on the marked stretch. 

The study showed that it is mainly at signal-controlled intersections where the travel time variation occurs. The recommendation was to invest in a signal priority system for about 50,000 euros instead of building bus lanes for 8-9 million euros. Bus lanes would not have solved the problem. 

This is a clear example of how data-driven insights can lead to more accurate and cost-effective decisions in public transport. By understanding where the bottlenecks are, we can prioritise the right actions - and avoid unnecessary investments.