Finding the weak points: New method to prevent train delay cascades

Train delays are not only a common annoyance for passengers but can also lead to significant financial losses, especially when trains are delayed across the rail network. When trains are delayed, it often sets off a chain reaction that turns small problems into massive delays across the system. This can be costly. A report from the Association of American Railroads (AAR) shows that rail disruptions nationwide could cost the economy more than $2 billion a day. The pressing question facing rail operators is therefore: how to effectively manage the knock-on effects of delays with the minimum of effort?

Researchers from the Complexity Science Center (CSH) used a novel web-based approach to quantify the systemic risk posed by individual trains to Austria’s entire rail network. “This allows us to identify weak links in the system, i.e. those trains that transfer significant delays to subsequent services,” explains CSH’s Vito Servedio. The study was published in npj sustainable traffic and transport.

Identifying the “Influencer Train”

The researchers built a network model by analyzing data from 2018 to 2020 on the route from the busy Vienna Central Station to Wiener Neustadt, which runs up to 1,000 passenger trains per day, as well as additional data on all train routes across Austria over a 14-day period. . In this model, nodes represent train services and links represent interactions that can cause delays. Using this model, the researchers were able to rank trains based on their potential to propagate delays and identify “influential trains.” To validate their findings and evaluate delay mitigation strategies, they built an agent-based simulation of Austrian railways that replicated daily train dynamics and interactions.

The results showed that trains running before and during the first rush hour were the most critical – “perhaps not surprising. But we can distinguish which trains are most influential in the intricate network of connections during rush hour, ” said Simone Daniotti, a doctoral student at CSH and first author of the study.

Additionally, the team observed that the risks associated with these trains are rooted in their intended dependencies. Only when a disruption occurs does the critical nature of these dependencies become apparent.

Rolling stock is the main cause of cascading delays

The researchers found that delay cascades in the model were primarily caused by shared rolling stock (locomotives and freight cars), even though there are fewer contact points between rolling stock than between infrastructure. “We see that materials such as rolling stock and people play a more important role in propagating delays across the rail network than the trains themselves,” explains Dagnoti. For example, if a train scheduled to depart at 2 p.m. relies on 8 a.m. rolling stock used by trains departing from the same point, any delay to the earlier train will seriously disrupt the later train. This creates a hard constraint that can be highly destructive.

While the current model does not account for staffing changes due to lack of data, it is designed to readily incorporate other factors such as staffing. This flexibility will allow for more precise analysis of latency impacts when these data points become accessible.

additional train services

To explore potential solutions, researchers simulated a one-hour delay for the busiest 2% of trains on the Austrian Southern Railway line from Vienna Central Station to Wiener Neustadt. These trains are considered to have the greatest impact on the network. “We found that adding three additional train services to the model reduced overall delays on key dates by approximately 20%,” Servedio explains.

The researchers say applying this approach across the entire Austrian railway system could reduce delays in the model by 40% and add 37 new trains or routes. They also observed that the more passenger traffic a railway line has, the more difficult it is to optimize.

Because the most cost-effective train services for rail companies to add are local trains with electric traction, and long-distance trains are more difficult and costly to replace, the researchers examined whether the different effects depended on the added train service. . “Interestingly, we found that by adding the three most cost-effective train services to the Southern Rail line, we could reduce overall delays by about 20%,” Servedio said.

pioneering approach

“Punctuality is one of ÖBB’s main goals. The model developed by CSH provides us with additional tools to achieve this goal in a complex railway system,” said ÖBB Project Manager Aad Robben-Baldauf.

CSH said: “Simulating a national rail system is very complex, involving large numbers of trains and operating points, and generating billions of scenarios. Traditional methods often fall short at this scale, but network analysis and complexity science provide powerful modeling tools to identify system vulnerabilities. This study exemplifies the significant benefits of linking scientific research with industry expertise, demonstrating how collaborative innovation can provide effective solutions to complex operational problems.

About the study

The study “Systemic Risk Approach to Mitigating Delay Cascades in Railway Networks” by S. Daniotti, VDP Servedio, J. Kager, A. Robben-Baldauf and S. Thurner was published in npj sustainable traffic and transport.

The study is part of the “Train Operations Forecast” project, a joint initiative between CSH and ÖBB that aims to develop optimization strategies for ÖBB’s passenger traffic to reduce overall annual delays on the system.