Delaware’s low elevation, crowded beaches, and limited exit routes make the state highly susceptible to severe flooding, and officials are hopeful that an infusion of federal infrastructure funding will help implement artificial intelligence-based evacuation plans in the future. On Thursday, the Biden administration planned to announce $53 million in grants to eight states, including Delaware, to develop high-tech solutions to traffic congestion problems. While these funds are part of the infrastructure law signed by President Biden in 2021, many of the programs, including the $5 million designated for flood response efforts in Biden’s home state, have evolved since then.
According to Shailen Bhatt, the former transportation secretary of Delaware and current US Federal Highway Administrator, the latest feature is predictive analysis and machine learning, which is now accessible due to an abundance of data. Bhatt explained that it’s difficult for human beings to differentiate between data and actionable information.
Delaware officials are accustomed to implementing evacuation-type procedures every week during the tourism season, with long lines of vehicles traveling to and from the beaches. However, flooding poses a unique problem, including standing water on roads, which can make the most direct town exits more dangerous than sheltering in place.
“What you don’t want to do is make the decision too late and then you have vehicles caught out,” said Gene Donaldson, who holds the position of operations manager at the 24-hour Transportation Management Center of the state.
The transportation department in Delaware is responsible for managing over 90% of the state’s roads, despite having the lowest average elevation in the United States. One of their major challenges is creating effective evacuation plans for high water events, which can be complicated due to the rapidly changing conditions. George Zhao, the transportation director for BlueHalo, a Virginia-based company that collaborated with Delaware on software development, acknowledged that it would be overwhelming for humans to monitor thousands of detectors or data sources.
This is where AI plays a crucial role. Instead of dispatching a team to obstruct a road that cannot be crossed, the technology employs sensors to identify potential weather hazards, and can even anticipate them. The system then transmits this data to drivers via mobile notifications and electronic road signs simultaneously.
As the number of automated cars on the roads continues to rise, the volume of data being generated is also increasing. These cars are not only able to alert their drivers of potential dangers, but they can also share this information with other cars on the road to prevent accidents.
One example of how this data is being used is the flood prediction analysis system developed by researchers at Missouri University of Science and Technology. The system was tested on the Mississippi River between 2019-22 and an earlier version was used. According to Steve Corns, an associate professor of engineering management and systems engineering who co-authored the study, the system was able to detect potential flood risks in a matter of minutes, a process that used to take hours.
Some potential functions of the flood prediction analysis system could include:
- Monitoring water levels and weather patterns to predict the likelihood of flooding
- Analyzing historical flood data to identify patterns and trends
- Providing real-time alerts and notifications to emergency responders and the public
- Identifying areas at particular risk for flooding and providing recommendations for mitigation strategies
- Supporting decision-making for emergency management, infrastructure planning, and resource allocation.
However, according to Corns, the current capabilities of flood prediction analysis systems are even more advanced and valuable than before, as long as they continue to receive adequate funding so that technology doesn’t become obsolete.
Past legislation had provided over $300 million in congestion relief grants, and according to Bhatt, the agency had received $385 million in applications for the latest $52.8 million batch under the infrastructure law, which indicates a significant demand for innovative solutions to tackle traffic issues.
In addition to the funding for the flood prediction analysis system, there were other payouts in the latest round of grants from the agency. These include $14 million for machine learning traffic prediction and signal timing in Maryland, as well as $12.7 million to retrofit the traffic system in Ann Arbor, Michigan, with cellular technology that could potentially be used as a template nationally. The grants also included $11.6 million for the expansion of a microtransit service in Grand Rapids, Minnesota.