June 20, 2012
What happens when you wait at a bus stop and the bus doesn’t show up? What if the same thing occurs five times or maybe ten? Do you give up on transit and drive? Try a different bus line? Buy a helmet and start biking?
Remarkably, researchers know relatively little about how reliability affects the transportation choices people make — even though it’s one of the most important factors affecting mode choice as well as popular support for transit.
ITS Ph.D. student, André Carrel, is trying to find out, and his efforts earned him a coveted Eisenhower award this month. The Eisenhower Transportation Fellowships enable students to pursue advanced degrees in transportation-related fields by contributing to tuition and living costs. Fellows also receive funding to attend the annual Transportation Research Board conference in Washington, D.C.
Carrel was drawn to this area of research after interning with the rail operations control of Transport for London. He observed how dispatchers reacted to disruptions in subway service, but discovered little if any useful information on how passengers reacted to these negative events. He wondered how many passengers transit systems lose as a result of unreliable service.
“The behavioral models we currently use almost never take reliability into account,” said Carrel. “Partly that’s because it’s very hard to observe, and partly it’s because passengers' decisions tend to be emotional rather than rational.”
He decided to take a different approach to understanding reliability’s impact on ridership by combining insights from behavioral economics with information gathered from mobile sensor and smartphone technology.
Working with Assistant Professor Joan Walker and Associate Professor Raja Sengupta, Carrel has developed a cloud-based infrastructure to automatically collect travel data over long periods of time from volunteers’ cell phones as well as real-time transit information from NextBus to see how unreliability on San Francisco's Muni lines alters passengers’ mode choice and to what extent.
“By connecting people’s trips to data from Muni lines we can see what effect certain events, as observed from an operational standpoint, have on passengers’ behavior and on mode choice.”
Ultimately, he hopes that this information will help transit agencies better understand how late trains and buses affect ridership and allow them to weigh costs and benefits of changing their operations.
“Preliminary results show that even a single negative experience can drive people away from transit,” said Carrel. “So, for instance, if dispatchers have to take a bus out of service because of a shortage of personnel, and riders decide to abandon Muni due to this bad experience, does it make financial sense to keep an extra driver on hand all day — which is costly even if he’s not used — or risk losing those riders?”
Perhaps it’s no surprise Carrel should find this subject so compelling: He grew up in Switzerland where public transportation is famous for running on time.
He’s hopeful his research will eventually help transit operators in the U.S. improve service — and with it their bottom lines.