Incident Echo Predictor

Inspiration

The inspiration for Incident Echo Predictor emerged from a recurring real-world frustration when trying to cross the bridges over the Tagus River. Even when navigation systems indicated low traffic after an accident, congestion would already be severe by the time drivers arrived, leaving them trapped with no viable alternatives. This gap between reported traffic conditions and what actually unfolds on the road highlighted a fundamental limitation of current traffic systems: they are reactive rather than predictive. The idea was motivated by urban mobility challenges and the need to support both drivers and road operators with earlier, more actionable information.

What it does

Incident Echo Predictor is designed to anticipate traffic congestion before it becomes visible on the road. The system provides predictive insights and route suggestions by forecasting how congestion caused by an incident is likely to propagate through the road network over time. While initially focused on specific incidents such as accidents, the concept can be extended to other types of disruptions. In the future, the solution is envisioned as a combination of a dashboard for operators and a mobile application for drivers, or as an integrated predictive layer within existing navigation platforms such as Waze or Google Maps.

How we built it

The project was developed as a conceptual prototype using Lovable to communicate and structure the idea. Rather than implementing a fully functional system, the focus was on clearly explaining the concept, logic and potential impact of predictive congestion modeling. The prototype demonstrates feasibility through simulated scenarios and conceptual data flows, illustrating how incident-driven congestion could be predicted and visualized across a road network. At this stage, the solution is intentionally conceptual, serving as a foundation for future technical development.

Challenges we ran into

One of the main challenges was the technical complexity of building such a system, particularly given the lack of experience in programming advanced predictive and graph-based models. Additionally, time and skill constraints required accepting that the initial prototype would be a basic MVP, not yet at the technical or functional level originally envisioned. Balancing ambition with feasibility was a key challenge throughout the project.

Accomplishments that we're proud of

The most significant accomplishment is reframing traffic management from a reactive approach to a predictive one. Instead of responding to congestion after it forms, Incident Echo Predictor focuses on anticipating what will happen next. This forward-looking perspective clearly differentiates the project from existing navigation tools, which mainly react to current traffic conditions rather than forecasting future congestion caused by incidents.

What we learned

Through this project, we learned that popular navigation platforms such as Waze and Google Maps are primarily based on real-time and historical information, not true future prediction. This reinforced the value and novelty of predictive congestion modeling. The project also deepened our understanding of how traffic systems behave as complex networks, where localized incidents can rapidly generate large downstream effects.

What's next for Incident Echo Predictor

The next step for Incident Echo Predictor would be to receive support in building a functional version of the system. This initial version would include the core features of the prototype: a visual map of the road network with incident locations, basic predictions of congestion propagation over time and preliminary route suggestions for drivers. Once a working system is in place, it could be expanded to integrate real-time traffic and incident data, improve prediction accuracy and provide more advanced decision-support for drivers and operators. The goal is to move from a conceptual prototype to a usable tool that demonstrates the system’s potential and value in real traffic scenarios.