Hi all,
I am part of a team of MIT Sloan students working on a class project analyzing MBTA elevator and escalator reliability, with the goal of predicting outages to support maintenance planning and improve accessibility for riders. We'd love to use real MBTA data rather than reconstruct it ourselves, and we're hoping someone here can point us in the right direction.
Specifically, we're looking for:
This is for an academic project only; we don't plan to redistribute the raw data and are happy to adhere to data usage standards or to credit the MBTA appropriately in our writeup. We're also glad to share our findings back with the team if useful.
Thanks so much for your time, and for maintaining such an accessible API — it's made this project possible even before this ask!
Thank you,
Mariana Vega
MBA Candidate at MIT Sloan
Hi Mariana,
Thanks for reaching out and looking to make MBTA-related data part of your project!
While the PredictionLoc tool mentions an internal archive that is largely deprecated at this time, we do have a number of datasets publicly available via our Lamp platform: performancedata.mbta.com.
One of the files listed on that page, `LAMP_RT_ALERTS`, is a long-term archive of our alerts data, including, but not limited to, elevator and escalator closures. The corresponding data dictionary can be found here. The archive is a Parquet file, for which various tools exist online to convert into a form that can be more easily viewed and analyzed.
Please feel free to reach out, either via deve...@mbta.com or in this forum if you have any questions about the archive's format and/or the interpretation of the alerts data!
Thank you,
Developer@mbta