Throughout the 1980s, Cassandras—who now heads the University’s Division of Systems Engineering—was deeply involved with what are called discrete-event systems. Computers, for example, are such systems, ones that obey rules that humans establish rather than rules that emerge primarily from physics and chemistry and other factors of the exterior world.
“Somebody came to visit me from Otis Elevator,” says Cassandras of the prompt that first launched him on a different trajectory. “I knew nothing about elevators, other than sometimes I would get very frustrated waiting for one to come. So, they asked me if I was interested in looking at how to schedule elevators.”
Knowing only a little at first about how such devices worked, Cassandras entered a world of what amounted to vertical traffic management design. His task: to refine a system so that elevators could better “decide” how to deliver the greatest number of riders to the most floors in the fastest manner.
These days, he’s working on systems that deliver a different kind of rider to a different kind of destination: a sensor-driven system aimed at helping drivers find parking spots on busy city streets.
“We use sensor networks in Smart Parking,” he says, referring to a municipally focused project in which wireless devices tell travelers whether a parking spot is reserved for them, open to a new reservation by them, or already reserved by another driver. The idea is that the system would also direct drivers to the best alternatives when it comes to parking near a particular location.
He and his team recently created a demo of the prototype system at a BU parking garage on Commonwealth Avenue. Next steps involve working with companies that are exploring wireless parking applications, and with cities and towns that might be interested in building such a network.
“It fits this broader vision of a smart city,” Cassandras says. “Imagine sensors like that being deployed for other purposes, detecting, for example, the status of power lines.”
