MIT researchers have developed an algorithm to identify the most efficient and cost-effective routes for Boston Public School’s fleet of 650 buses; other software lets school districts create flexible pickup and dropoff schedules. It’s part of a growing effort to use artificial intelligence (AI) to get people places faster. “The name of the game is optimization,” Virtual CTO Vaclav Vincalek told Lifewire in an email interview. “You need a system that can adjust itself in real-time based on traffic conditions, so the traffic lights can improve the flow and favor public transportation, school buses included.”
Getting Kids to School Faster
Boston public school staff used to take weeks to build bus routes, but MIT’s solution devises routes in approximately 30 minutes. MIT researchers used data from Google Maps to analyze traffic patterns during morning and afternoon rush hours. They used mapping software and optimization techniques to devise an algorithm that would reduce the number of bus routes, reconfigure bus stops, maximize the number of students riding each bus, and cut the time empty buses are on the road. They also considered that some students require wheelchair-friendly buses, and others need home pickup. “This speaks to the power of optimization and large data sets,” Dimitris Bertsimas, an MIT professor, said in the news release. Another program called TravelTracker is used by schools to plan better routes. It automates the process of creating bus routes and lets parents and staff know when buses are running late. Derrick Campbell, director of transportation for the Del Norte Unified School District in northern California, said that the software makes bus routes as safe and efficient as possible.
AI Eases Traffic
School buses aren’t the only vehicles getting places faster, thanks to AI. In Germany, researchers are working on a model that would speed up pedestrian intersections by up to 15 percent. The system uses LiDAR sensors and AI to measure the speed at which the pedestrian crosses the intersection, ensuring the pedestrian isn’t left in the middle of the crosswalk as the light changes. Companies like Uber or Lyft are already using systems to optimize the number of cars on the road and their routes, too, Vincalek pointed out. But one of the bigger benefits is how AI traffic control could even save lives. “Suppose through AI a real-time navigation system makes a routing decision during a heavy rainy day that results in less travel time, less congestion, less chance of an accident, and ultimately, higher quality of life,” Stevens Institute of Technology AI & transportation expert Yeganeh Hayeri told Lifewire in an email interview. “In that case, we can say an AI-directed decision making has defeated an individual’s shortsightedness resulting in better mobility and accessibility.” In an email interview, Jai Ranganathan, chief product officer of the transportation firm KeepTruckin, also backed the idea that AI can make roads safer. KeepTruckin produces AI dash cams that can monitor the road ahead and the driver’s behavior, providing real-time alerts that improve safety. The AI can also automatically follow up with driver coaching to improve their safety record over time. AI could go a long way to adjusting a couple of unfortunate statistics. For one, the fatality rate for the first nine months of 2021 increased to 1.36 fatalities per 100 million vehicle miles traveled. Beyond that, truck drivers are responsible for moving 72 percent of all the goods we consume, and 74 percent of all fatal passenger vehicle cases include a large truck. “As a result, our customers can reduce their number of accidents by 36% on average,” Ranganathan said. Another use of AI in transportation is predictive maintenance, Ranganathan said. “Think about the times you’ve seen the “check engine” light flash in your car or alerts for when you need to change your oil,” he added. “Using AI as predictive maintenance software to gauge potential failures and scheduled repairs increases your vehicle’s lifespan and reduces operating costs.”
