For municipalities that haven’t moved over to all smart cards for their buses, or have a combination of tickets, paper-based passes, and electronic boarding, it’s been hard to estimate accurate ridership numbers for particular routes and times of day, in real-time.
Head counts are one method, but are expensive and require an individual to perform the task. In-bus infrastructure, such as turnstiles, can be linked to location services such as GPS and analyzed, but that’s complicated and any hardware needs maintenance. Again, more cost.
Researchers reckon they have a solution. And it’s one that they say can be brought in for about $60 per bus.
The idea is to simply capture all of the MAC addresses of all of the smartphone users on the bus.
University of Washington transportation engineers say in a paper that they’ve developed a method of capturing MAC addresses as way to find out “where bus riders get on and off, how many people use a given stop and even how long they wait to transfer to another bus,” an article on the university’s website says.
A MAC address, or Media Access Control address, is a network identifier used by the smartphone for communicating data. It’s often just on anyway.
UW’s system can estimate passenger origin and destinations from Wi-Fi radios when they’re broadcasting and Bluetooth radios when they’re discoverable.
“Say you have a Husky game or Seahawks game and you want to know how much demand changes so you can offer the right level of bus service,” authorYinhai Wang, a UW professor of civil and environmental engineering and director of thePacific Northwest Transportation Consortium, said in the article.
“If you can gather enough data from these real-time sensing systems, that’s going to offer very valuable information,” he said.
Sensors like this have been used to gather vehicle travel times on roads, but this is one of the first times it’s been used on buses, the researchers say in the article.
By collecting the time and location of the signal too, they reckon that they can get accurate numbers.
Other road users
The researchers say that their numbers are good in part because they’ve written an algorithm to go along with their radio-collecting tool.
Through a time-on-board calculation, the algorithm can exclude other road users, such as a pedestrian walking alongside the bus. That person isn’t alongside long enough—in Seattle traffic—to be mistaken for a passenger.
How accurate can it be?
A possible hiccup, though, is what about senior citizens? Are they as likely to be carrying a media-intensive, MAC address-spouting phone as a Seahawks-attending Millennial? Then there are economic factors. Some regions may contain more smartphone-wielding passengers than others.
As of April 2015, 64% of American adults owned a smartphone, according to Pew Research Center. However, ownership levels “remain particularly low among seniors, as just 27% of Americans 65 and older now own a smartphone,” Pew says on its website.
The researchers say spread is not a problem. The article says Kristian Henrickson, a co-author on the paper, claimed the penetration of cellphone ownership will help the benefits outweigh the drawbacks.
“Think about understanding how long and disconnected a route may be from some less-privileged neighborhoods to an employment center,” Henrickson said in the article. “This technology provides a much better way of assessing that and possibly improving upon that.”