Encouraging citizens to use public transport is a major issue for urbanized areas around the world. It reduces congestion on major road arteries, provides a mode of transport for those who don’t drive or can’t afford to buy a vehicle and facilitates access to services and facilities in other locations.
Singapore has one of the most efficient public transport networks in the world with a Mass Rapid Transit (MRT) network of under and above ground lines linked with Light Rail (LRT). From all accounts it’s a well-utilised service but there are still issues, as in most major cities, with congestion during peak hour periods.
Mr Lew Yii Der, Group Director of Corporate Planning and Research at Singapore Land Transport Authority (LTA) said earlier this year “Even as we expand the rail network and bring in additional trains to enhance rail capacity, it will take some time for these improvements to fully come on stream. Travel demand management is an area LTA has been actively exploring as it encourages more optimal use of public transport capacity.” Part of this exploration has been the launch of a Singapore-based project to study trends associated with travel on public transport. But more interestingly the study is also looking at how incentives and game mechanics can reduce congestion during peak periods.
Insinc with commuting congestion
Insinc (Incentives for Singapore’s Commuters) is a joint collaboration between the National University of Singapore, Stanford University and LTA to study the effects of incentives and web-based interaction on public transit commuting. Since January, the program has rewarded commuters for traveling outside of peak periods. Participants register their EZ-Link card so that each journey earns credits (1km = 1 credit) while shoulder-peak travel (6.30am-7.30am and 8.30am-9.30am on weekdays) earns bonus credits (1km = 3 credits). The study is the first of it’s kind in Singapore but has well-established roots in several projects that have successfully encouraged behaviour change.
Pedometers, Indian traffic jams and campus parking
Professor Balaji Prabhakar, Professor of Electrical Engineering and Computer Science at Stanford University is head of the Insinc study and his ground-breaking work on traffic congestion has been well documented in publications such as the Economist, NY Times and Wall Street Journal.
Prabhakar’s academic expertise allows him to appreciate the complex relationships inherent in public transport networks. However, he is quick to point out that such networks are all around us. “Societal networks don’t have to do with computing or communications, they’re all the other networks we use on a daily basis. Technology and economics interplay strongly and humans are also in this loop. We are components of the network, so our decisions matter, while we are in the system. Whether we choose to take the MRT, a bus, bike or car, it affects the load within the system… The key issue is supply and demand but how do you make them meet? Price can be used, as economics suggests, but could we also use incentives? That’s the purpose of our work.”
The study of incentives to encourage behaviour change led Prabhakar and his team to turn to the world of video games and gamification (using game elements in non-game contexts). Last year the team collaborated with global management consulting giant Accenture US to develop Steptacular, a pilot study that encouraged employees to get fit through the use of pedometers, earning and then redeeming credits for rewards. Results indicated “high enrollment, frequent engagement with the system, and an improvement in the average steps per user per day”. Even professional game designers from RedOctane, developers of legendary video game Guitar Hero, have commented on the design and noted similarities with well-known gaming phenomena such as “level grinding” and the “Easter eggs of random rewards”.
But it was a visit to Bangalore and getting stuck in peak hour traffic that revealed to Prabhakar the potential intersection between societal networks, tracking behaviour and creating feedback loops to modify future behaviour. A resulting study involving 14,000 users, successfully shifted 25 percent of peak hour congestion, resulting in the removal of eight buses from the existing fleet. The most telling result was that the financial savings from those extracted buses was more than enough to cover the incentive rewards used in the program.
Calculating the Odds
Capitalizing on the findings of these previous studies, the Insinc program uses several game elements and mechanics to encourage behaviour change. Commuters earn credits and bonus credits depending on how often and when they travel. Additionally, the number of “decongestion” trips determines the commuter’s “level” within the program. This is where things get interesting. Insinc commuters can exchange their accumulated credits for cash (credited back to their EZ-link card) or win larger prizes by playing simple games of chance on the Insinc website. As the player “levels up” they gain access to boards with increasingly higher stakes to win larger amounts.
Insinc utilises elements from common graphical user interfaces (GUI) found in-game designs to communicate odds and chance. As Prabhakar says, “This idea of visually explaining things to people, where it becomes self-explanatory, is one of the things we’ve taken very seriously. It underlies a lot of the game design that we’ve been doing.” A recent addition to the game tool set has been the “Magic Box”. These weekly offers are designed to encourage specific behaviour within small groups of users allowing the developers to trial different types of incentives and measure their effectiveness.
We’re all part of the network
When modifying behaviour one needs to take into account the overall ecosystem that surrounds the individual. A range of factors including work, study and family commitments can affect the timing and frequency of public transport habits. Obviously it’s not possible for everyone to travel earlier and in fact, it’s not even desirable. For example, if everyone started traveling earlier you would have only just shifted the congestion: same problem, different time. Consequently the aim of the study is to encourage just a proportion of commuters to alter their behaviour and only then, by just a few days in the week by 20 to 25 minutes.
Although Insinc has been in operation for little over six months it has already produced some impressive results. Around 22,000 commuters have registered for the program with over SGD180,000 paid out through the system to date. But has the study been able to modify the public’s commuting behaviour? The answer is yes: of those commuters who have participated in the program, almost ten percent of their peak period travels have shifted to the shoulder-periods.
Hardly surprising then that LTA recently announced that the study will be extended a further 18 months to study the effectiveness of the scheme over a longer term and across a wider cross-section of commuters, anticipating the addition of 40,000 new users. As of August 1, Insinc will be opened to commuters with NETS Flashpay cards, senior citizens and student concessions (universities, polytechnics and Institute of Technical Education.)
The gamers reading this article will probably be wondering: if you’re going to introduce game elements why not introduce an element of strategy? Prabhakar notes that it’s a question of skill versus chance. In the case of Insinc, the use of game mechanics provides an engaging interface while the element of pure chance simply provides a random reward redemption option. The point is not to beat the system but to work with it. “There are more interesting things we can do with social elements,” Prabhakar says, “for example where friends can play together. That makes it more engaging.”
Earlier this year Prabhakar launched a three-year project based at the Stanford campus in California aimed at reducing the congestion of cars arriving and departing campus during peak periods. Known as CAPRI (Congestion and Parking Relief Incentives) the study uses RFID technology to monitor car movements and then reward drivers who traveled during off-peak periods. Given the success of previous projects, stakeholders will be eager to track the results of the project.
The use of game elements in Insinc has provided concrete results of how incentives rather than disincentives can successfully modify behaviour when it comes to public transport adoption. However, I believe the program’s success also demonstrates a rapidly growing trend. This type of system, one that provides the user with access to her own data, is becoming prevalent due to the advances in mobile technology and sensors as well as the increasing appetite of users’ to track, reflect and modify their behaviour. It’s self-awareness guiding self-improvement. And if we can harness this powerful idea to change our public transport commuting habits… imagine what else we can change.
This blog post first appeared on http://www.recognitionpattern.com