In this article we discuss how cities can enhance their smart city systems by incorporating a behavioral approach – a design layer that leverages their hardware, infrastructure and data to promote behaviors that are in alignment with civic goals. Without these behavioral enhancements, smart city systems are unlikely to solve key urban challenges—in travel demand management, resource management (energy and water consumption), waste management, public services and safety—or improve urban efficiency and quality of life.
The concept of smart cities has been all the rage in urban planning circles for some time. Local governments attempting to improve city efficiency and technology firms eager to sell hardware and software solutions have wholeheartedly embraced and promoted the concept. No single definition exists for what a “smart city” should consist of. Generally, however, it includes the deployment of physical Internet of Things (IoT) devices and sensors connected to information and communication technology (ICT) platforms, or the utilization of disaggregated data from mobile phones or other devices to allow cities to collect and analyze data to manage resources. The idea is to use this data to help cities address urban challenges in transport, resource and waste management, governance and public services and safety.
Many cities around the world have to varying degrees adopted these technologies and are already utilizing them, but in spite of the large-scale efforts and considerable resources invested by cities like New York, Rio de Janeiro, Amsterdam and Madrid and by countries carrying out nationwide efforts like India’s Smart Cities Mission, to date, most of these systems have little to show for in terms of improved urban processes or quality of life.
The problem with smart city systems as they have been conceived thus far is that they emphasize monitoring, data collection and analysis and almost exclusively focus on supply-side policies: e.g. automatically reduce energy supply during non-peak hours, optimize traffic light systems to prioritize public transit, alert garbage collectors of waste disposal gluts. Although these efforts are necessary and certainly an improvement from past conditions, they do nothing to address individual behaviors that exacerbate urban challenges: How do we encourage individuals to reduce energy and water consumption at home? How to do we nudge commuters to use public transit or non-motorized vehicles rather than private automobiles? How do we reduce the production of waste in homes and businesses? Without addressing the demand side of urban challenges (i.e. human behavior), we are unlikely to solve the critical challenges that cities everywhere face today.
The data that is created by smart city systems can and should be operationalized to promote individual behavior change, but our current systems are lacking the behavioral design layer that can allow them to nudge individual behavior into alignment with citywide objectives. This critical lack of a behavioral perspective embedded within existing smart city systems ultimately makes them useful at monitoring, collecting and centralizing data, but renders them incapable of modifying collective behavior and truly improving urban processes and quality of life. So how can we leverage these technologies to actually change behavior and improve our cities?
Behavioral Smart Cities
Today, we have the opportunity to leverage these smart systems to actually modify behavior by incorporating a behavioral design layer. Ultimately, what this means is creating parallel behavioral smart systems that go beyond monitoring, collecting and analyzing data and that effectively operationalize the information smart city systems produce to promote behavior change. These parallel systems can be anything ranging from simple infrastructural design elements to more complex apps and communication algorithms that interact with residents.
By way of example, let’s consider traffic congestion: most large cities around the world today face tremendous traffic congestion problems that only seem to get worse over time. While congestion pricing has worked to some extent in cities like London, Singapore, Stockholm and Milan, others have found it difficult, impractical or impossible to institute similar incentives-based systems. An alternative to congestion pricing is travel demand management, which can consist of a series of processes that nudge individuals to modify transport-related choices: Do I commute to work in my car or do I use public transit? What time do I leave my house? Which route do I take? The aggregation of the thousands (or millions) of individual choices in response to these questions ultimately determine the level of congestion in a given city. So, how can we influence those choices so as to minimize congestion?
Currently, through the use of smart city cameras or decentralized traffic data collected from mobile platforms (e.g. Google Maps or Habidatum), cities around the world possess real-time data that allows them to obtain a relatively good snapshot of current traffic conditions. And, yet, few cities are operationalizing this data to promote behavior change. A hypothetical example of this “behavioralization” would be to leverage information from IoT systems and individual travel pattern information voluntarily shared by citizens, to send commuters personalized “active” push notifications with suggestions on how to commute to work, not only to optimize their individual personal route, but with the aim of reducing overall city congestion. Here’s an example:
Push notification/SMS sent at 7:45am: “Good morning, John. Consider taking the SunRail train at 8:29am from the Tupperware Station in Kissimmee to the Church Street Station in Downtown Orlando to save 20 minutes and $4.78 on your morning commute. For detailed directions, click here.”
This message leverages the data obtained from the IoT systems and delivers it in a way that nudges residents to modify behavior by employing several behavioral techniques:
1. Appropriate moment of choice: the message is delivered at the moment in which John is about to make his decision, prior to leaving his home. This can be achieved by employing his personal historical commute pattern data and his GPS location.
2. Salient medium of delivery: the notification is delivered to his smartphone, via a push notification or SMS that is likely to get his attention.
3. Choice simplification: the behavioral algorithm already did the homework for John and is telling him exactly what he needs to do; he doesn’t have to actively search what is the best option for his morning commute and detailed instructions are also provided.
4. Added value and immediacy: by providing John with estimates of time and monetary savings, the behavioral message creates a perception of value that motivates him to take action. This knowledge provides an immediate reward to John that allows him to internalize future and social gains.
All of this can be achieved through a simple app that, given prior consent and appropriate data-protection safeguards, could easily create a customized experience that ultimately nudges John to modify his behavior in alignment with broader urban goals: namely, lessen traffic congestion on the Florida Turnpike, but also decrease emissions in the Orlando Metropolitan Area and reduce the need for maintenance and investment on infrastructure for automobiles. Thus, rather than only using smart city data to optimize traffic lights, we could also employ the information that these systems yield towards modifying individual behavior
A real-life example of a smart city system that can be further enhanced to maximize its behavioral potential to reduce water consumption is New York City’s Automated Meter Reading. This system consists of a radio transmitter connected to a water meter that sends daily readings to a computerized billing system. With this mechanism, consumers are able to monitor their consumption on a regular basis. How might we behavioralize this system so that it goes beyond monitoring and actually encourages consumers to reduce consumption? Some starting points:
1. Create a two-way communications platform (e.g. app) that allows consumers to not only check their consumption, but to also receive alerts via SMS or push notifications.
2. Use social norming comparisons to reduce water consumption by showing consumers how their water consumption compares to that of their neighbors.
3. Establish an algorithm that alerts consumers when they are close to exceeding their three-month average or their neighborhood’s three-month average.
4. Alert consumers in case an AI algorithm detects a leak or a faucet that was left on.
These are only some examples of what might be done. To fully maximize the potential of these smart city systems, a thorough behavioral diagnosis can be carried out to determine how to best enhance these smart city systems so that they may modify consumer behavior based on agreed-upon behavioral objectives for the city.
Behavioral Design for Smart Cities
Ultimately, what is required to transform smart city technology into a system that is able to change human behavior is a deliberate effort to design behavioral processes and systems that operationalize this data for behavior-change purposes. This enhancement requires a delimitation of the behaviors that the system will seek to address (travel choices, energy consumption, waste management, etc.) and a behavioral diagnosis for each behavior that allows the city to create a design strategy that nudges individuals to align their behavior with overall urban goals, whether that be reducing traffic congestion and emissions or encouraging the use of public services. Through iterative behavioral design processes and appropriate testing, cities can make the most of their investments in smart city hardware and data and ultimately improve quality of life. Without a behavioral lens, these smart systems are unlikely to live up to the expectations of the citizens that are footing the bill.
Interested in learning more on how you can use Behavioral Science to improve your city? Need support in optimizing your smart city systems? Our real estate and transport specialists at the BVA Nudge Unit would be delighted to help. Contact us.
Héber M. Delgado-Medrano is a Behavioral Design Consultant and currently serves as the VP of the BVA Nudge Unit in the United States. He is passionate about improving cities and transport and loves to ride electric scooters, e-bikes and his V2 Dual+ Boosted Board.
This is the first in a series of blogposts on using behavioral design to improve the sharing economy. Electronic […]