You’ve probably read in countless places that cities are booming like never before in the history of civilization as centers of residence and commercial activities. It’s also probably clear to most anyone reading this article that there are myriad ways that technology can be harnessed to improve city operations and quality of life for city dwellers. But it may be less apparent what it takes to implement those new technologies and what challenges slow the pace of progress.
By Amie Devero
The explosion of change in digital technology is quickly expanding in its applications to urban issues. So whatever successes exist are just the tip of an iceberg that, frankly, is mostly unknown. Estimates suggest that Internet of Things (IoT) spending on devices and services will exceed $1.7 trillion by 2020 , according to IDC. We can imagine some of the ultimate intersections of the cloud, the Internet, ubiquitous wi-fi connections and the Internet of Things –a universe made of objects that have connectivity built into them. For example, it wasn’t so very long ago –only 10 years— when parking meters were always mechanical. I’m sure you recall fishing in your car’s ashtray for spare change to purchase time at a meter by turning a steel knob after inserting coins into a slot. Then, even multi-space parking kiosks (the big boxes that let you pay for a space with a credit card), were somewhat novel. Today, mechanical parking meters are anachronisms that exists only in places where the city either relies so little on parking meter revenue, or has so little revenue that it doesn’t pay to replace them.
So the most basic level of Internet of Things devices are probably the very parking meters you use every day: Credit-card accepting, license plate tracking and wifi-connected machines that can tell a cloud-based database when the space is occupied, for how long and by whom. Their occupancy data and payment status can be accessed by parking meter enforcement officers using handheld devices similar to mobile phones. The enforcement devices also print citations –and can double-check before printing one that a space hasn’t been paid for in the last few seconds– reducing errors by officers and the time those errors take up later in traffic court and beyond. Those same parking enforcement devices are connected to other means of payment and other data sources besides the parking kiosks. Now that parking can be paid for digitally, with an app, PayPal, a credit card (attached to an app) or through Near Field Communication (NFC), an example of which is Apple Pay™. When a parking enforcement officer checks his device to see if a specific license plate has paid for the space it occupies it checks more than the parking kiosk. The device also checks whether it has been paid for by a mobile app, and whether the vehicle has been moved from a nearby space to avoid time limits on parking. Finally, it will also check whether than license plate number has been included on a list of vehicles that are tagged for towing, booting, amber or other alerts. If the car comes back as (for example) to be booted, the enforcement device can automatically request a tow-truck or booting vehicle be dispatched.
All of that is happening at the small, local level of enforcing a single vehicle in a single parking space. But there is more that the meter kiosk can do besides ensure that the municipality collects all of it parking revenue. Because the meters have information about each space and whether it is used, and a record of when it was used and for what amount of time and money, something so basic as parking space data can add up to a mass of information. That information can be collected, aggregated, parsed, sliced, diced and analyzed. And the information that results can inform a city engineer of what locations get the most use, how much money could be raised by staggering parking rates, where the traffic jams are likely to be from people circling to find a space and so much more. All that from parking kiosks. The data can then be accessed in any number of ways by various other types of services. For example, there are plug-in apps available that will provide an end-to-end service from the calendar on your phone, through your navigation app, your parking payment app and the event you are attending. Imagine you are scheduled to attend a concert taking place tomorrow at 8 PM. You have put that in your Google calendar at your PC. Your mobile phone is linked to your Google account including your calendar. At approximately 7 PM tomorrow, you receive an alert from Google that tells you it is time to think about leaving for the concert. The time of the alert was determined by another app running in the background, one called Parking-Polly ™ by Spark Parking Technology. Parking-Polly has checked the parking occupancy of the area around the concert via both parking devices and human information (crowd-sourcing) and determined where there are some spaces available. You have previously logged a preference for spaces that are close to the venue, with a secondary preference for getting the lowest price possible. Based on that information, Parking-Polly is directing you to leave at 7 PM, because the cheapest and closest spaces will be on a street that is filling up quickly. Parking-Polly plugs into your navigation service, and directs you to the Arena via the street where the parking spaces are most likely to be available. You follow the direction of Google to leave at 7 PM, and navigate directly to the most reasonably priced parking space for the concert. And then, of course, you pay for the parking space using your parking mobile payment app.
This kind of a seamless, “frictionless” experience is the vision of every smart cities enthusiast. But the whole experience I just described is still confined to the parking function provided both by the city and private parking operators. The service never touched any other silo within municipal function. Think about all of the other municipal functions that you interact with regularly and you’ll see what I mean. Public works departments take care of the infrastructure of city life. They handle sewers and roads, streetlights, police, potholes, trash collection and much more. The information that didn’t inform your app could still create problems for you on the way to the concert. What if you wanted to park and ride the bus instead of parking? You would have needed the bus timetable to provide that information to Parking-Polly or to Google. Or, what if there had been an accident on the road and the police had closed it down so there was no parking? Imagine there had been torrential rain in the preceding week, and the roads had standing water that was treacherous –but you didn’t know that Parking-Polly may not know that the reason behind the availability of parking spots is that the road is impassable.
Of course, if there were flooding and potholes, we would also want notification of those challenges to go not only to you, the customer, but to the appropriate city functionaries who are responsible for those problems. And there are services available today that do that. They are called digital 311 services (311 is the telephone number established for municipal repair requests in the US). Those services provide a means for a city-dweller to report problems to the city. In Bridgeport, Connecticut if you see a dead animal in the road you can use the BConnect app (customized by QScend ) to photograph the road kill, GIS locate it and send it directly to the waste management department. The report will be time-stamped and logged with the map of the location, and you can track the progress of it via the app. If the report is not closed as complete within a pre-configured period of time, the report will escalate to the Supervisor, and after the next increment, to the Director, and finally, if unresolved, to the Mayor. But again, while digital 311 services are irreplaceable, they too exist in isolation from other city functions. They do not necessarily provide data to anything other than the department responsible for fixing the problem at-hand. So the dead animal will get picked up and properly disposed of, but the street light that made the street dark and therefore led to the raccoon not being seen by the oncoming truck may not get reported. Moreover, if the animal died of natural causes, there is no methodology for sending the necropsy information to the animal services or health department so that they can leap into action to avert a rabies epidemic, or notify wildlife commissions of an encroachment by wild game into a neighborhood.
Of course, technology exists to do everything I described and much more. In fact, there is enough technological progress to essentially have every single part of a city talking to every other part. Today, there are already many cities where a great number of IoT technologies are already monitoring services, human activity, health information and environmental conditions to save energy, add efficiency and make like easier. For example, there are cities where moisture in the soil is monitored and the sprinklers calibrated only to water when it is needed, thus saving water. Cities can add sensors to streetlights that check for movement and dim when none is present to reduce electricity usage. In Boston, when traffic is bad, highway signs provide estimates of the time that a motorist can save by pulling over to the nearest train stop and jumping on a train –reducing traffic congestion and automobile emmisions. That information is available from roadside signs, but also by requesting it from a mobile phone. Or in Seoul, S. Korea, where there is mass density (45,000 people per square mile) the city has placed GPS-enabled touch payment units in its 25,000 taxis. By virtue of the GPS tracking, data about taxi locations and routes is collected and can be used elsewhere. The payment system is a boon for customer efficiency, but the real bonus is the data about congestion, parking and the predictive utility that other city services can utilize.
A holistic, comprehensive system
There is much more, but providing it is simply a laundry list. What’s noteworthy is that they are all discrete from each other. So in a single city there may be multiple instances of “smart city” technologies functioning with the use of IoT objects, but they are not part of a holistic, comprehensive system. Each individual government silo within the city has procured a unique solution to solve a single problem or set of problems. The only places that have begun it look at a bigger picture are either cities that are sufficiently small they could build systems from scratch, or newly formed “smart cities’ that may or may not even have a significant population benefiting from the integrated services on offer.
To a point, this piecemeal approach is fine. The various jobs get done –the water saved, the carbon footprint reduced, the parking faster and simpler. But from a slightly more long-term perspective, it will turn out to be a significant obstacle to fulfilling the potential of IoT in smart cities and will be a monumental systematic challenge down the road. To understand why it can become problematic we only need to look at the technological challenges that all kinds of organizations, including cities, deal with now by virtue of having allowed technology to evolve naturally (and thoughtlessly) rather than with significant and long-term planning. In the private sector where, in almost every large organization, there have been mergers, acquisitions, spin-offs or simply silo-ed management, integration teams deal with significant mismatches of technology and data between accounting, finance, payroll, HR and other systems. The fact that these systems can’t “talk” to each other is a very expensive problem in terms of money, time, and human frustration. That’s the same problem we see in cities. It’s completely ordinary for a single city to require multiple custom technology integrations to accomplish the most basic tasks throughout a value chain. For example, it’s common to need several special work-arounds simply to get the information about the status of parking citations into the computer system running the traffic court –and to require yet another integration to convert the data in the traffic court to populate the finance system where the revenue is counted. Crazy, right?
Why is it done this way instead of through a comprehensive plan? Well, there are lots of reasons. The most obvious of course is that there are only a limited amount of financial resources—money—at any given time. So projects are discrete and limited of necessity. This is particularly stark in the US where municipal coffers are always dwindling for a variety of reasons, including the shadow of the recession, the lack of middle-class incomes to support high sales tax revenue, the fact that state politicians consistently lower taxes and with those reductions, cut back on the formulas that provide state funds to cities, and the increasing demands made on cities to support new and growing needs as their populations increase.
But that is the least problematic of the explanations for the non-systematic approach to smart city-building. The more insidious issues have to do with the culture and structure of government planning and procurement. In local government, especially in the US, functions are highly decentralized. So the visionary leader within the municipality –who may be the mayor, city manager or even a council chair– has little direct impact on the departmental procurement of new technology. Each functional silo of the municipality is learning, changing and buying on its own, without any likely direct feedback or partnership with other departments. So, for example, unless the Mayor himself champions the digital 311 service, it may only apply to public works. Or the parking department may be charged with increasing revenue and therefore procure new citation equipment and service -but fail to account for the collection of code enforcement revenue, health citation revenue, court costs and scheduling, or business licenses and so forth (all of which should probably be rolled into the same system). With each department planning and procuring independently, there is very little attention given to the eventual need for interdependency.
Moreover, if there is a visionary leader driving progress –for example, a mayor who is current and keen on smart city technology, she has her own concerns that will necessarily inform how she approaches the project. Hers (and most mayors) is a political role –elected or appointed— and she must manage her own popularity –if only to ensure she is around long enough to bring the project to fruition. The requirement to be popular leads to some odd consequences. Popularity comes from “sexy” changes; changes that provide immediate pain relief or efficiency to customers. For example, constituents love quick and ready apps that make parking easier, bus schedules more predictable or parks more fun. But the real infrastructure building that a comprehensive technology plan would require might be slower. Laying groundwork is ponderous and can be hidden from public view for long periods of time. The interstate system, the New York City subways and the London Underground were not quick projects. They took decades. And they exerted a toll in inconvenience and cost. Those kinds of projects are popular when finished, but politically onerous to accomplish.
Few parallel strategic themes
So the challenge that emerges is how to do two things at the same time: add smart city solutions on an incremental basis that fit into limited budgets, while at the same time: build a system that will support organic growth, modular additions and agnosticism –while providing enough “sexiness” along the way that constituents will tolerate cost and inconvenience. It’s not a simple problem. One way to approach it is to work on a few parallel strategic themes. On the one hand, to develop a long-range systematic and coordinated plan –20 years out at least. And within that plan, create milestones for integrations and coalescence of functions –EVEN if it isn’t clear today what those functions may be.
The secondary parallel theme must be to address all systems with each infrastructure initiative. For example, create a procedure for pulling in all departments -at least to listen, weigh-in and provide feedback– whenever new initiatives are under consideration. This may or may not lead to actual collaboration, but it could. And if the feedback is enthusiastic enough for one or another initiative, it may lead to starting a project with two or more functional silos involved from the outset.
One thing some cities have done is to lay down a framework of infrastructure groundwork without necessarily knowing all the places it will lead. That can mean simply installing sensors on everything –roads, parking spaces, streetlights, water meters, electrical meters, in soil and municipal buildings– even without connecting them to anything but a monitor. But the existence of that infrastructure begins a context shift that will bring about uses for the data and new ways of connecting data to other services. An ancillary project would be to provide access to all data to the public, even in those early days of collection. That opens the door to the innovators who have ideas that no one else has yet conceived. They can use that data, connect it, dissect it, parse it, model it and repackage it in ways that will bridge gaps in our current knowledge and imagination. There are cities doing all of these things now, and they will gain traction and a head start over other cities because of those choices. For example, using sensors and surveys, Portland State University and the City of Portland, Oregon, have collaborated to collect air quality data along corridors where a major transit line begins operating in 2019. They are also collecting data on the effects of that new transit corridor on the character and air quality of the neighborhoods it goes through. It’s not entirely clear right now what the usefulness of that data will be. I suspect it will be handy though.
In Chicago, a new project called “The Array of Things” will start by collecting data from the environment, roads, transit and more. As part of the project they have placed sensors in as many locations as is possible. The collected data is available to the public at large. The hope is that ‘[t]he data will help make Chicago a truly “smart city,”’ and will allow predictive modeling for flooding, other weather catastrophes as well as being used by new applications to track air quality and provide alerts to city-dwellers.
Of course, the best possible way to approach the project of building smarter cities using IoT is to build a long-term plan, fund it and begin now. There are places doing that, but they are not usually older cities with the complexity and layers of organic growth, huge and diverse populations and myriad interlocking systems that many of the most populated and historic cities have. But we are in early days. For now, even these side-doors and tunnels into the massive scope available for municipal IoT are important and worth watching. I’m keeping my eyes glued to the space, and I encourage you to do the same.