AI Usage at a Glance
Jun 18, 2020
Data AnalysisPractice documented: The MBTA uses a machine-learning system, developed in partnership with vendor 4Atmos Technologies, that analyzes oil samples taken from diesel commuter rail locomotives to predict engine failures 30 to 60 days before they occur. The program began as a pilot in spring 2018 and transitioned to full production across the commuter rail locomotive fleet.
Practice DocumentedView practice →Jun 27, 2024
Customer SvcPractice documented: The MBTA tested a generative AI chatbot developed by Northeastern University students that assists MBTA customer service representatives at The RIDE paratransit service by quickly retrieving policy information and answering frequently asked questions during live calls with riders. Customer service staff began using the chatbot during testing in 2024; as of the most recent sources it remained an internal tool for representatives, not a public-facing system.
Practice DocumentedView practice →Nov 1, 2024
ProductivityPractice documented: The MBTA integrated AlphaRoute's NDSP Optimization tool in 2023 to automatically determine each day which paratransit trips should be assigned to dedicated vehicles versus non-dedicated service providers such as Uber and Lyft. In 2024, the system optimized over 1.24 million trips, saving an estimated $22.88 per outsourced trip and reducing scheduling staff's manual work by 4–5 hours daily.
Practice DocumentedView practice →Nov 12, 2024
ProductivityNew evidence: MBTA, AlphaRoute Hit 1M Optimized Trips on Paratransit System
Evidence AddedView practice →Dec 17, 2024
ProductivityPractice documented: The MBTA integrated Spare's AI-driven platform into its paratransit service, The RIDE, to power real-time trip scheduling and dynamic routing for riders with disabilities across Greater Boston. The new system launched on August 30, 2025, replacing legacy scheduling software, and the MBTA confirmed it as its official transportation management system via a contract signed in December 2024.
Practice DocumentedView practice →Dec 19, 2024
Customer SvcNew evidence: Northeastern at the forefront of Massachusetts' AI-driven economic growth, governor says in announcing AI Hub
Evidence AddedView practice →Jan 27, 2025
Data AnalysisPractice documented: The MBTA deployed a machine-learning system developed by vendor LYT that predicts when a bus will approach each intersection, enabling Boston's traffic management center to extend green lights for buses in real time. A proof-of-concept test on Brighton Avenue in Allston, running from July 2024, reduced red-light wait times by 21% and cut overall travel time by 8% on routes 57 and 66; the MBTA and City of Boston announced plans to expand the system citywide in January 2025.
Practice DocumentedView practice →Jan 29, 2025
Data AnalysisNew evidence: Boston upgrades traffic signals to improve bus travel times and reliability
Evidence AddedView practice →Jan 30, 2025
Data AnalysisNew evidence: Faster bus commutes in Boston? That's the goal with coming traffic signal upgrades.
Evidence AddedView practice →Feb 13, 2025
Data AnalysisNew evidence: How the MBTA Turned to AI to Improve Bus Travel Times in Allston Corridor
Evidence AddedView practice →Sep 4, 2025
ProductivityNew evidence: The MBTA's The RIDE reimagined with Spare, powering 1 million trips a year
Evidence AddedView practice →Sep 9, 2025
ProductivityNew evidence: Transforming Paratransit: AlphaRoute & MBTA win METRO's Innovative Solutions Award
Evidence AddedView practice →Sep 16, 2025
ProductivityNew evidence: MBTA Launches 'The RIDE – MBTA' Mobile App, New Online Portal for Paratransit RIDE Customers
Evidence AddedView practice →Sep 19, 2025
ProductivityNew evidence: Boston's MBTA, AlphaRoute Redefine Paratransit Efficiency with NDSP Optimization Breakthrough
Evidence AddedView practice →Nov 21, 2025
Data AnalysisNew evidence: The MBTA is using AI to make the system run smoother, one oil sample at a time
Evidence AddedView practice →The MBTA tested a generative AI chatbot developed by Northeastern University students that assists MBTA customer service representatives at The RIDE paratransit service by quickly retrieving policy information and answering frequently asked questions during live calls with riders. Customer service staff began using the chatbot during testing in 2024; as of the most recent sources it remained an internal tool for representatives, not a public-facing system.
The chatbot was built by Northeastern University students in the AI for Impact Co-op Program as part of the Massachusetts InnovateMA government partnership. It takes natural-language queries from MBTA call takers as input and searches MBTA policy documents and websites to produce cited answers that representatives can relay to callers. The tool also generates lists of sources for further reference. The developers and Mass.gov confirmed the tool was being used by MBTA customer service representatives during testing in mid-2024 and was highlighted at a briefing with Governor Healey. As of the most recent sources it is not confirmed to be a permanent production deployment; the developers noted it could potentially be adapted for external use in the future.
The MBTA integrated Spare's AI-driven platform into its paratransit service, The RIDE, to power real-time trip scheduling and dynamic routing for riders with disabilities across Greater Boston. The new system launched on August 30, 2025, replacing legacy scheduling software, and the MBTA confirmed it as its official transportation management system via a contract signed in December 2024.
Spare's platform receives inputs including vehicle location, vehicle requirements, customer wait times, and service performance metrics, and produces optimized real-time trip schedules that adapt to unexpected conditions such as traffic or customer emergencies. The AI also powers a consumer-facing mobile app and online booking portal for The RIDE, launched September 2025, giving riders real-time vehicle tracking and booking flexibility. Before this transition, The RIDE operated on legacy software that did not offer real-time optimization tools, and staff dealt with manual workflows. The MBTA officially confirmed the system change in a press release and reported that on-time performance rose from 88% to 99% after launch.
The MBTA integrated AlphaRoute's NDSP Optimization tool in 2023 to automatically determine each day which paratransit trips should be assigned to dedicated vehicles versus non-dedicated service providers such as Uber and Lyft. In 2024, the system optimized over 1.24 million trips, saving an estimated $22.88 per outsourced trip and reducing scheduling staff's manual work by 4–5 hours daily.
The AI tool receives daily trip data including service agreements, customer eligibility criteria, and capacity limits, and produces an optimized daily assignment plan routing each eligible trip to the most cost-effective transportation option. Before the tool, MBTA scheduling staff manually reviewed trips and used spreadsheets to determine which to outsource to non-dedicated providers—a process AlphaRoute documented as requiring hours of labor per day. The tool was selected by the MBTA in 2023 and rolled out from pilot to full scale. AlphaRoute is a company with MIT ties and uses MIT-developed routing algorithms.
The MBTA integrated Spare's AI-driven platform into its paratransit service, The RIDE, to power real-time trip scheduling and dynamic routing for riders with disabilities across Greater Boston. The new system launched on August 30, 2025, replacing legacy scheduling software, and the MBTA confirmed it as its official transportation management system via a contract signed in December 2024.
The MBTA integrated AlphaRoute's NDSP Optimization tool in 2023 to automatically determine each day which paratransit trips should be assigned to dedicated vehicles versus non-dedicated service providers such as Uber and Lyft. In 2024, the system optimized over 1.24 million trips, saving an estimated $22.88 per outsourced trip and reducing scheduling staff's manual work by 4–5 hours daily.
The MBTA deployed a machine-learning system developed by vendor LYT that predicts when a bus will approach each intersection, enabling Boston's traffic management center to extend green lights for buses in real time. A proof-of-concept test on Brighton Avenue in Allston, running from July 2024, reduced red-light wait times by 21% and cut overall travel time by 8% on routes 57 and 66; the MBTA and City of Boston announced plans to expand the system citywide in January 2025.
The MBTA uses a machine-learning system, developed in partnership with vendor 4Atmos Technologies, that analyzes oil samples taken from diesel commuter rail locomotives to predict engine failures 30 to 60 days before they occur. The program began as a pilot in spring 2018 and transitioned to full production across the commuter rail locomotive fleet.
Have evidence about Massachusetts Bay Transportation Authority (MBTA)'s AI practices? Submit a report.
Submit a report →AI Trace is free and nonprofit. Support our work