We’ve all been there. The time comes to leave the big game, Black Friday shopping, or some other event that draws a crowd, and everyone is left shuffling their feet due to the inevitable congestion. Fujitsu wants to change that and has begun field trials on a smartphone app that gives incentives to those who would wait it out. Using an artificial intelligence-enhanced system it calls Human-Centric Zinrai, the app aims to find the best candidates for staying behind and the incentive most likely to entice them to do so.
The app comes out of Fujitsu Laboratories, an R&D outfit established in 2014 in Singapore's Centre of Excellence, and was developed in partnership with the Agency for Science, Technology and Research and the Singapore Management University.
NEW ATLAS NEEDS YOUR SUPPORT
Upgrade to a Plus subscription today, and read the site without ads.
It's just US$19 a year.UPGRADE NOW
The goal of the partnership is to research congestion in the real world, gather data on how it forms and develops, and come up with solutions to relieve congestion in various circumstances. Focusing on urban areas, the research so far has not only mapped alternative routes for pedestrians, public transportation, and traffic, but has focused specifically on incentives-based options for moving people away from the congested area.
The app’s goals are to both show those entering congested areas how to avoid them and to keep congestion from happening in the first place by mitigating its effects. To that second end, Fujitsu and partners have put a lot of research into what types of incentives, using dollar figures as measurements, are most likely to move certain types of people away from congestion or potential congestion.
Research has suggested certain models that are likely to match those willing to modify their departure date or movements and the incentives that are most likely to get them to do so.
In Fujitsu’s behavioral guidance models, three key points are involved: satisfaction level, behavioral-guidance factor, and receptivity to suggestions. Each works into a formula for likelihood of a specific incentive triggering the desired response. Satisfaction level is a measurement of the person’s tolerance to congestion/crowds, how much they can delay or advance departure, and how much they’re willing to pay. Behavioral-guidance factors are a measurement of how much satisfaction levels change when incentives are given. These incentives can include transit discounts, coupons to nearby businesses, and similar. Receptivity is the likelihood that a given incentive is going to modify behavior.
So far, survey data has shown that about 51 percent of people will spend more time at a shopping center as a means of avoiding congestion when notified of it. A coupon worth US$3.50 will entice about 73 percent of people to stay and wait and a coupon worth twice that will entice about 91 percent of them to wait. Those survey results are the basis of the framework for Fujitsu’s calculations.
The results could provide alternatives to relieving congestion, which currently rely on infrastructure expansion such as new roads or rail lines. Those survey results, for example, point towards a congestion point of about 10,000 people being reduced by 30 percent if 40 percent (4,000) of those involved were enticed away from the congestion point temporarily. This, in turn, would add that much more traffic to nearby businesses.
Fujitsu and its partners began field testing on November 1, which will continue at several congestion-prone facilities throughout Singapore until December 2017. The artificially intelligent software will learn along the way, hopefully improving as it goes. Test results will not only help advance the technology and be used to create the application itself, but also models for city engineers and businesses to use globally as a means towards reducing congestion. A commercial version of the app may enter the market as early as spring of 2016, built into Fujitsu’s real-time location data analytics cloud service, SPATIOWL.Source: Fujitsu View gallery - 3 images