Clever elevator system saves energy

Ding! Researchers have come up with a way to make elevators more energy-efficient — by tracking potential passengers with smart building sensors.

Scientists have done plenty of studies to improve elevator scheduling, but most have aimed only to shorten the users’ waiting time. For example, one system counted the number of passengers waiting for the elevator; if there were too many people to fit into one car, it sent multiple cars. Other researchers have proposed installing buttons for each floor next to the elevator (instead of simple up or down buttons) so that passengers can select their destination floors in advance.

The authors of the current study wanted to reduce both wait times and the elevators’ energy usage. Consider this scenario: One person is waiting on floor 16 and another on floor 18, and they both want to go to the first floor. If the second person hits the elevator call button too late, the elevator may have already picked up the person on floor 16 and be on its way down; it’ll have to make a second trip up afterward. But if the system could detect the person on floor 18 in advance, the elevator might be able to pick up both passengers in one trip.

So the researchers proposed using building sensors to detect potential passengers before they even press the elevator call button. They outfitted a 21-story building with devices such as video cameras and floor sensors. Using those observations, they could detect people coming toward the elevator about a minute in advance and “reserve” the elevator for them. About 84 percent of the people who triggered elevator “reservations” ended up pressing the elevator call button.

With this system, wait times could drop by 15 to 30 percent and energy usage by 28 to 31 percent, the team estimates in Building and Environment. Next, the researchers hope to find a way to figure out passengers’ destination floors in advance as well. Roberta Kwok | 12 December 2013

Source: Kwon, O., E. Lee, and H. Bahn. 2013. Sensor-aware elevator scheduling for smart building environments. Building and Environment doi: 10.1016/j.buildenv.2013.11.013.

Image © Kiefer pix | Shutterstock