How a Rush Hour Commuter Disruption Could Lead to Sustainable Resilience
This past Friday, BART, one of the primary mass transit systems in the San Francisco Bay Area experienced unplanned downtime during the morning commute. BART averages 360,000 riders a day, 150,000 of whom cross the Bay and would have been affected by the delay. Yet, by 9am, ridership was only 30,000 below normal. Clearly, most of the passengers didn’t get the message.
Communications of the downtime were shared over traditional media outlets and commuters were advised to work from home, carpool, or take advantage the additional buses that were put in service from other transit agencies. Coordination among transit agencies fared much better than coordination between transit agencies and their end users. Compared to service disruptions for terrorist threats, earthquakes, or other natural disasters, this was a very minor incident. Even so, it did underscore the need for much better communications.
Now, imagine a scenario where you’re a regular BART commuter on one of the affected lines and when you wake up, you have a text message, voicemail, and e-mail notifying you of the unplanned downtime along with alternate suggestions and instructions on where to find the latest information.
This is all possible today at ever-decreasing costs to implement. Using noSQL or real-time databases (e.g., Hadoop, Cassandra, MongoDB, Neo4J) greatly improves the ability of finding the affected users based on historical usage patterns and could do so in a matter of minutes or hours as opposed to the technologies of a few years ago that would take days or weeks to process such a query, if it could be done at all. This data analysis could be further expedited by in-memory computing (e.g., SAP HANA, Oracle Exadata). According to Gartner, the use of in-memory computing technology will rise three-fold by 2015.
We live in an era of relationships at scale – where organizations can now have 1-to-1 relationships with end users while simultaneously being able to communicate with all of their customers. A good example of this is the National Weather Service’s integration with Facebook and Apple that notifies geographically select individuals of impending disruptive weather events with no prior opt-in required. Overall, the data shared with digital services is increasingly relevant to more stakeholders, increasingly accessible through APIs, and increasingly ripe for big data analysis.
Today, a vast majority of commuters in the San Francisco Bay Area use a Clipper Card as a single form of payment across multiple transit agencies. Just tag your Clipper Card on a bus or turnstile and payment is made. For auditing purposes, Clipper also keeps track of when and where each payment occurs and Clipper has an e-mail address and phone number associated with each account.
After the couple hours it might take to extract the relevant data from a real-time database, it can be shared with a Reverse 911 system through an API call. Reverse 911 systems have been deployed in cities throughout North America and are capable of placing landline phone calls to geographically select individuals; newer systems can send SMS messages and e-mails as well. With the BART collision happening at 2:35 a.m., calls, texts, and e-mails could start flowing before most people get out of bed—which should be plenty of time to boost that 30,000 number to something much more resilient.
We may not live in the age of the Jetsons, but there’s no reason our public transit infrastructure can’t live with us in digital harmony. With the technology available today, public transportation can become resilient infrastructure where downtime is effectively communicated in a scalable way. After all, if a city wants to be resilient, then it must also be digital.