This story is part of Governing's annual International issue.
The city of Santander is an old seaport on the north coast of Spain. A couple times a week, a ferry brings over a handful of English tourists who crowd the beautiful beaches of Playa de la Magdalena and El Sardinero, and shop and dine in the city’s historic center. Otherwise, the Spanish city of 180,000 has little interaction with the foreign world. Or at least that was until Santander was chosen four years ago to become Europe’s test bed for a sensor-based smart city.
Today, the sight of foreign tourists is a familiar one. Delegations from tech companies the world over are flocking to “see” Santander’s sensors. Since 2010, 12,500 sensors have been placed in and around the city’s downtown district, where they measure everything from the amount of trash in containers, to the number of parking spaces available, to the size of crowds on the sidewalks. In addition, sensors on vehicles such as police cars and taxicabs measure air pollution levels and traffic conditions.
The data from these sensors flows to banks of computers that analyze the real-time information and give city officials the kind of big picture that allows them to adjust the amount of energy they use, the number of trash pickups needed in a given week and how much water to sprinkle on the lawns of city parks.
At the same time, the city is opening up its data so that programmers can create apps that help citizens find bus arrival times or let tourists find out who is performing at concert halls simply by pointing their mobile phones at a bus stop or building. The collecting of data through numerous sensors could lead to significant improvements in how city infrastructure is used and lead to a better understanding of urban issues, says Carlo Ratti, director of the SENSEable City Lab at the Massachusetts Institute of Technology. “Sensors allow us to better understand reality,” he says, “letting us catch the pulse of a city.” The goal in Santander is to make the city a more enticing place to visit, shop and get around. These kinds of sensor-driven projects are already saving the cash-strapped Spanish city significant amounts of money.
But others are more skeptical about the rush to try to measure as many data points as possible in a city, run them through analytical software and then base major urban policy decisions on what the data says. “I think there are some considerable risks that we are building into cities by deploying these technologies very rapidly, somewhat haphazardly and without a lot of risk assessment,” says Anthony Townsend, a senior research scientist at New York University and author of Smart Cities: Big Data, Civic Hackers, and the Quest for a New Utopia. Townsend and others have raised concerns about buggy software and the push to optimize aging infrastructure through the use of sensors. Where some see smart cities, others fear a new form of governance based on algorithms rather than human experience.
While the idea of a smart city is still relatively new in the United States, the concept is more established abroad. London, Seoul and Stockholm have been using sensors to monitor traffic and manage congestion for years. Singapore has placed sensors throughout almost every part of its physical geography with the goal of becoming the world’s smartest city. And Rio de Janeiro has become the poster child for smart cities with its high-tech central operations center, which is staffed by nearly 400 workers who monitor everything from traffic to keywords in local social media in an effort to spot trends -- or problems -- before they occur.
Four years ago, the European Union (EU) decided to use Santander to test the latest in smart city technology. Dubbed SmartSantander, the project, which received an $11 million grant from the EU, started when Luis Muñoz, an engineering professor at the local University of Cantabria, and his development team of 20 technicians, researchers and programmers buried hundreds of sensors in the asphalt to develop a way to manage the limited number of parking spaces available in the city’s downtown area.
Parking management, says Muñoz, is a good place to start. It is one of the low-hanging fruits as far as testing sensor technology goes, because it gives drivers a quick solution to the nagging problem of where to park and helps city officials reduce congestion and air pollution. Information about parking is displayed on special panels located at major intersections in the city, so anyone who is heading downtown will have an idea of how many spaces are currently available and where they are located.
After that initial project, the team soon launched others, including adding 2,000 sensors on streetlights and buildings to monitor temperature, carbon monoxide levels, noise, light and traffic; another 60 sensors to monitor traffic flow into the city; 50 sensors in two major city parks to measure environmental factors; and so forth.
Muñoz’s team wrapped up the project last year, and turned it over to city officials who use all of these monitoring and sensing systems to more precisely calibrate how the city allocates its resources. The sensors on the streetlights, for example, can tell whether the sidewalks have people on them or whether the streets have cars. If it’s late at night and the streets and sidewalks are empty, the lights are dimmed. If the sensors pick up someone walking, they brighten. These sensors and others have helped Santander cut energy costs by as much as 25 percent. Similarly, sensors in trash cans have helped the city cut garbage pickup costs by 20 percent.
But sensors installed by technicians aren’t the only way Santander is testing the smart city concept. “People with smartphones can be an even better instrument for broadcasting information about the city and facilitating new interactions with it,” says MIT’s Ratti. He points to Waze, a community-based traffic and navigation app that shares real-time traffic and road information with users. It is as an example, he says, of how people can become sensors themselves.
Santander has launched an app called “Pace of the City,” which users can download to receive up-to-date information about city events. The alerts are even geocoded so that people can use their phones to get to events. Geolocation plays a prominent role in what Muñoz calls “augmented reality” apps. Using about 2,000 quick response (QR) codes -- those square, checkered symbols that can be scanned with a smartphone -- the city has tagged points of interest, shops and public places around Santander. They have proven useful as a way to provide tourists with information about what the city has to offer and where they can find it. Geographical location, for instance, is also helping with the city’s transit system. Using one of Santander’s apps, a person can get bus arrival times simply by pointing the phone at the bus stop.
The more the public uses the city’s data in these applications, according to Muñoz, the better they get at interpreting the data and what it means. “For example, when they see one of the displays saying there are five open parking spaces in a certain location in the city, they know that the information is correct only up to a point,” he says. “They know the actual number may change.” So when the number of spaces posted on the display fall below a certain number, drivers can decide whether they should chance it and search for a space, or find other alternatives.
Certainly at this point everyone is excited about sensors and their possibilities. But a growing chorus of experts is calling for a more measured approach, arguing that there are broader and deeper problems when cities become too reliant on sensors and smart applications to drive services and decision-making. As cities increase their dependence on software to run services, operate infrastructure and make critical fiscal decisions, they increase the risk that something could go wrong, says NYU’s Townsend. He cites Y2K as the best example of when software can backfire: It cost companies and governments $300 billion to fix. The Israeli government has acknowledged that essential services that run off sensors, such as water, electricity and banking, have been the target of numerous hacking attacks. In 2012, the traffic management system for a main artery in the port city of Haifa, Israel, was hacked, causing major traffic problems that lasted for hours.
Then there’s the so-called brittleness factor. A good portion of the sensors being used in cities are designed to improve the performance of key infrastructure, such as roads, rail, water systems and electrical grids. Townsend describes this as squeezing excess capacity out of existing networks. “You’re creating a structure that is inherently unstable and can only be controlled by a computer and software that can sense what’s going on,” he says. When these systems are forced to operate faster and to constantly recalibrate their settings based on demand, the potential for a breakdown increases. “So what happens when these control systems shut down?” asks Townsend. “Does it happen gracefully? Does it fail catastrophically? Does it warn you? Are there digital backups?” The real-world analogy for this predicament is what happened to cellular networks during hurricanes Katrina and Sandy and the Boston Marathon bombing. Circuits overloaded and the networks shut down.
Professor Muñoz admits that’s a problem. With the sheer number of sensors that are spread throughout the city, “bad weather can create problems,” he says. “That can be difficult to manage when you have such a large, distributed network of devices.”
Ultimately, there’s also the question of privacy versus security. Sensors that capture sound, images and locations of individuals can be helpful when we need protection or greater security. But there is always the worry that what starts as surveillance in a democratic society can become something else in a country that becomes autocratic. Rob Kitchin, director of the National Institute for Regional and Spatial Analysis at the National University of Ireland, points out that the more a city can measure and control, the more likely it becomes “panoptic,” capable of tracking everything, never forgetting what happens. In other words, a kind of Big Brother society.
But spring in Santander means the surf is up and those who like to ride the waves are more interested in apps that can provide up-to-date information on wave size than privacy issues. Information about surf is another data point SmartSantander can share with visitors and the local population. Muñoz agrees that privacy is one of those issues that the project is trying to tackle. “We are concerned about what to share and what should remain private,” he says. “One of our goals is to encourage users to embrace this technology, not run away from it.”
Another goal is to make city government more transparent than ever before, Santander Mayor Iñigo de la Serna told Germany’s Der Spiegel. “We want to create a new, cooperative relationship between the people and the city government.”
One promising new application would leverage the sensors that measure sound so that when they pick up the frequency of an ambulance siren, the traffic control system kicks in and creates a priority lane of green lights so the ambulance can get to the hospital faster. That’s the kind of story Muñoz and de la Serna would like people to remember when they think about the possibilities of smart cities.