Improving Transport Planning Through Digital Twin Innovation Using SAS Viya

Challenge

Urban rail networks are increasingly vulnerable to disruptions caused by extreme weather events, technical failures, and large-scale public events. These incidents often lead to cascading delays, passenger congestion, and communication challenges for operators and policymakers. Traditional systems lack the capability to anticipate and model these impacts in real-time, hindering effective response and mitigation strategies.

Solution

Building upon our success as Global Grand Champions of the SAS Hackathon 2024, we aimed to push the boundaries of analytics and real-time data to transform the management of transport challenges. For our SAS Hackathon 2025 submission, our team developed a proof of concept for a Digital Twin of a regional rail network, built on SAS Viya.

The "Event Horizon" Digital Twin integrates live data streams with forecasting, optimisation, and intelligent decisioning models to replicate the behaviour of a rail network under stress. This system can:

• Predict disruptions caused by weather events, such as floods or storms, using live APIs and historical data
• Model the cascading effects of delays or closures across the wider network
• Suggest alternative routes or transport modes for passengers
• Forecast passenger volumes during events or peak travel times
• Automatically trigger alerts and rerouting workflows through Agentic AI

Our approach utilised multiple components of the SAS Viya platform:

• SAS Studio: Data preparation and processing
• SAS Visual Analytics: Real-time dashboard and visualisation
• Model Studio & Model Manager: Predictive and forecasting model development
• Event Stream Processing: Integrating live API feeds (e.g., flood warnings)
• Visual Forecasting: Predicting passenger numbers at peak times and during major events
• Optimisation: Identifying the most efficient alternative routes
• Intelligent Decisioning: Automating alert workflows and rerouting logic

For example, when a flood warning API triggers an alert, the system immediately identifies the high-risk area, recalculates the most efficient routes using optimisation models, forecasts the impact on passenger volumes, and automates rerouting decisions through Agentic AI.

Impact

The Event Horizon Digital Twin demonstrated significant potential in enhancing the resilience and efficiency of rail networks:

• Real-Time Disruption Prediction: Enabled proactive identification of disruptions, allowing for timely interventions
• Dynamic Rerouting: Facilitated the swift adjustment of services to minimize delays and passenger inconvenience
• Forecasting Capabilities: Provided accurate predictions of passenger volumes, aiding in resource allocation and crowd management
• Automated Decision-Making: Streamlined operations through automated alert systems and rerouting protocols

This proof of concept lays the groundwork for a scalable solution that can be adapted to various regional rail networks, enhancing their ability to manage disruptions effectively.

Building upon the success of the Event Horizon Digital Twin, we aim to collaborate with transport authorities to pilot this solution in real-world scenarios. By integrating additional data sources and refining predictive models, we seek to further enhance the system's accuracy and responsiveness. Our goal is to establish a comprehensive framework that empowers rail operators to anticipate, mitigate, and manage disruptions, ensuring a seamless travel experience for passengers. Please get in touch if you are interested in understanding more.