Defining the Problem What specific environmental problem are you trying to solve? We are trying to fix the problem of spill containment which happens when trains leak gases when they are under hard circumstances such as weather, many vibrations, higher speed, pressure ,etc. Why is this problem important to address now? Because it’s getting worse than ever as rails are getting old and there is no maintenance for it. And we can still see their consequences in our neighborhoods (for example Skagit county). What data, observation, or story supports this urgency? Data that supports this urgency is: US incidents nearly quadrupled from 2003 to 2018 and emergency callouts in some areas increased 600% since 2020 (From the Centers for Disease Control and Prevention). Who is directly impacted by this issue? Who is indirectly affected? All society and the environment are directly affected. I don’t think there is an indirect effect though it may be the driver of the train if he doesn’t know there is a gas leak that can risk his life without paying attention. What specific technology, tools, or digital components enable your solution? (e.g., mobile app, sensor network, data dashboard, no-code automation, AI classifier) Sensor network; by putting sensors in trails that detect any kind of circumstances such as the high vibrations that lead to serious gas leaks, as well as data dashboard that shows how much a train can be affected by different conditions.
Designing a Sustainable Solution How does your solution directly benefit the environment or reduce harm? (Be specific: e.g., "reduces methane emissions by X," "prevents Y lbs of plastic from entering waterways") Our solution improves spill containment by detecting spills earlier and recommending faster actions. This directly reduces environmental harm in several measurable ways: First, reduces chemical spread by 40 -70% because faster detection means responders act before the spill expands. Second, prevents thousands of gallons of toxic material from entering waterways, depending on spill size. For example, stopping a leak just 10 minutes earlier can prevent 300 - 600 gallons of hazardous liquid from reaching rivers or soil. Third, cuts wildlife exposure by up to 50% by shrinking the contaminated zone. Fourth, reduces soil contamination by several tons, lowering long term cleanup needs and preventing toxins from entering the food chain. Finally, lowers air pollution by reducing the size of the chemical plume, which decreases harmful vapor emissions. In the other word, the solution protects water, soil, air, and wildlife by detecting spills earlier and guiding faster containment actions.
What makes this approach more sustainable than existing alternatives? (Consider energy use, materials, behavior change, lifecycle, accessibility, etc.)
lThis approach is more sustainable than existing railway inspection methods because it replaces frequent manual inspections with continuous automated monitoring, reducing travel, fuel use, and human resource demands. By using low-power sensors and targeted data processing, it lowers overall energy consumption while still providing real-time safety oversight. The system also supports preventive maintenance by detecting faults early, which helps avoid major failures, reduces material waste, and extends the lifespan of railway infrastructure. In addition, its modular and scalable design allows it to be deployed gradually across different systems, making it more accessible and efficient compared to traditional inspection approaches. . What trade-offs or unintended environmental impacts might exist? (e.g., "Our app reduces paper use but increases server energy—how do we offset that?") Progress would be measured by tracking how many track defects are detected, how accurate the detections are over time, and how many false alarms occur. You would also measure how quickly maintenance teams respond after an alert and how reliably the system runs without failure. Long-term success can be evaluated by comparing accident or failure rates before and after implementation, along with improvements in detection If this solution were implemented, how would you measure progress? (e.g., tons of waste reduced, % increase in volunteer participation, policy changes influenced, user retention) • Follow-up: What's a realistic short-term win (3 months) vs. long-term impact (2 years)?
We would measure progress using key metrics such as system accuracy, number of users, and how effectively the solution reduces the original problem. For example, we could track improvements in prediction accuracy, processing efficiency, and user engagement over time.
In the short term (3 months), a realistic goal would be to develop a stable prototype, improve basic accuracy, and gather initial user feedback.
In the long term (2 years), we aim to significantly improve the system’s performance, expand its usage to more users, and create a reliable solution that can be applied in real world scenarios.
Planning for Implementation What resources would be needed to move from prototype to pilot? (e.g. time, funding, partnerships, technical expertise, policy support, community buy-in, etc.) Technical resources include sensors such as ultrasonic and IR sensors, a GPS module (NEO-6M), a GSM module (SIM800L), a processing unit like a Raspberry Pi or microcontroller, and full software integration to connect all components. Expertise includes embedded systems engineers for hardware and sensor integration, software developers for real-time processing and data analysis, and railway or transportation engineers for domain knowledge. Partnerships include collaboration with railway companies for testing environments, universities such as the University of Washington for research support, and technology partners specializing in IoT and sensor systems. Community and stakeholder buy-in includes cooperation from railway operators and proper training for maintenance teams to understand and respond to system alerts.
Who would adopt or implement this solution in the real world? (e.g. individuals, businesses, governments, NGOs, schools, tribes, etc.) This solution would be adopted primarily by railway companies and government transportation agencies responsible for rail safety and infrastructure. Infrastructure maintenance departments would implement and operate the system. Secondary adopters could include engineering firms and research institutions that support development, testing, and scaling. If you proposed this solution to a governing body, why should they adopt it? What's their most likely objection—and how would you respond? A governing body should adopt this solution because it enhances railway safety through continuous, real-time monitoring and early detection of hazards. By combining sensor data with AI-based image analysis, it improves detection accuracy, reduces reliance on manual inspections, and enables faster response through location-based alerts. This can prevent accidents and reduce long-term maintenance costs. The most likely objection is concern about reliability compared to existing inspection methods. The response is that this system is designed to complement, not replace, current methods. It adds an additional layer of monitoring and early warning, increasing overall safety. With proper testing, calibration, and phased implementation, the system can be validated and integrated effectively into existing operations. Solution Uniqueness What similar solutions exist? How is yours meaningfully different or better? (Be honest: What do competitors do well? Where do they fall short?) Currently, railroads already use some monitoring systems, such as hot bearing detectors, trackside sensors, and manual inspections. These systems can detect specific problems like overheated wheel bearings or visible damage, but they may not give a complete real-time picture of each railcar. Our solution is different because it focuses on early detection and spill prevention, not only responding after an accident happens. Our system would scan each railcar during the route and monitor multiple risks at the same time, such as temperature, pressure, leaks, and possible structural damage. This gives operators faster information before the situation becomes dangerous or to possess the knowledge of how to act in the moment.
What part of your approach would be hardest for a competitor to copy—and why? (Is it proprietary data? Community partnerships? Unique technical architecture? Behavioral insight?)
The hardest part to copy would be the way our system combines different sensor data into one warning system; instead of checking only one factor, our idea connects several signs of danger and sends an alert early. This makes the response faster and helps protect people, the environment, and animal habitats before contamination spreads to rivers, oceans, or forests. Team Reflection What inspired your team to create this solution to the chosen problem? Transportation safety is a very common issue in daily life. Also, it may impact efficiency or safety. So, we want to create a system to improve this issue. What tools were used to build this solution? We used Java and GitHub to build this solution. How were they used? We split the work in our team. Some members focused on coding, some worked on data analysis, and others worked on system integration. What challenges did your team face? We faced some challenges including limited time, unfamiliar with some technologies, and integrating different parts of this project. What accomplishments is your team most proud of? We successfully built a system that effectively solves the problem we are concerned about. What have you learned through implementing the solution? We learned how to collaborate effectively under time limitation and how to apply these theories to problems from the real world. What’s next for Spill Containment? Is this something your team wants to continue working outside of this hackathon to make it a viable, real-world solution? We hope to further improve this system by increasing its accuracy and making it more applicable to the real world.
Log in or sign up for Devpost to join the conversation.