Chicago in winter is gripped in the throes of ice and snow. Traveling around on the frozen streets is frustrating and dangerous for city drivers, until the path is cleared by one of the city’s more than 300 snow plows.
In the past, residents anxiously peered out their windows, watching to see when the plow would come down the street so they could unbury their cars and head out in the morning. But now, they can simply pull up a page on the city website and, on a clever animated map, watch the progress of each of those 300 plows in real time.
- Syracuse University - M.S. in Applied Data Science: GRE Waivers available
- SMU - Master of Science in Data Science - Bachelor's Degree Required.
- UC Berkeley - Master of Information and Data Science Online - Bachelor's Degree Required.
- Syracuse University - Master of Information Management Online
- Maryville University - Master of Science in Business Data Analytics
- Villanova Business - Master's in Analytics and Study Data Mining, Predictive Analytics Online
City planners have always sought more information about how the residents of their cities live, and today they are finding it. More surprisingly, they are seeing the benefits of giving that data back to the public, influencing and assisting city residents by showing them in detail and with clever visualization what effects they are having on the urban environment:
- Monitoring power consumption to find patterns that will help residents and businesses lower their bills
- Finding and distributing information on open parking spaces to drivers in real time
- Tracking fire hydrant status and distributing information to local residents, allowing them to shovel out snow-buried hydrants prior to emergencies
- Using GIS systems to plan zoning changes and keep citizens informed with easy visual presentations about building projects and progress
The synergy of data science and municipal planning will determine the future of the world’s cities, and it will all be in the hands of master’s-educated data scientists.
Seeing the Signs: Putting Big Data on Maps Makes Zoning and Maintenance Easier
Some of the oldest and largest data sets municipal planners use come from land use and zoning records.
Long before other factors in city living were recognized as important, planners understood that the physical layout of housing, commerce, and industry could be critical to the success of an urban center. Zoning, the practice of delineating what land uses were allowed in what areas, was one of the earliest applications of data to planning.
Today, zoning and land use are largely run on the backs of massive Geographic Information System (GIS) datasets. Combining topographic information with environmental, ownership, legal, and other data, GIS maps give planners an unprecedented ability to visualize a neighborhood- and to extrapolate the effect of zoning changes before they are made.
Sharing this data would allow municipalities from across the country to see what other cities have experienced so they can identify practices that have worked or failed elsewhere. Although a formal GIS data-sharing program between cities has yet to come, urban planners already spend time consulting public-access data sets that many cities and counties have made available.
GIS systems are being integrated with environmental data to analyze vulnerability to flooding and other natural disasters and adjust building codes in vulnerable areas to ensure safer standards. In Glendale, California, city managers integrated their GIS system with a street sign database and discovered that they had about 2,000 more street signs than it thought it had. At $200 each, that could well exceed the traffic department’s budget when it comes to replace them if this hadn’t been accounted for.
Keeping Traffic Flowing with Real-Time Data Visualization
In modern American cities, cameras cover nearly every intersection, in-pavement sensors count vehicles as they pass, and every traffic light reports status updates back to a central control center. Traffic maps, such as those offered by the Washington State Department of Transportation, have become common tools for planners and commuters alike.
Datanami, which covers big data applications in government and business, described how San Diego transit planners use a tool that incorporates tracking data from five separate data sources to fuse together multi-modal routes that commuters use each day. This has allowed planners to find the optimal ways to link together the city’s bus and trolley system to connect travelers for the fastest commutes.
In many other major cities, including Seattle and Boston, racks of bicycles stand on busy corners, available for bike-share program members to grab and ride for an hour or two from point to point for a nominal fee. In a control center in the city, managers watch for bikes accumulating at some spots and growing thin in others. Trucks are dispatched to rebalance the system when necessary. And apps make the information available to foot-sore pedestrians who are looking for the closest available bike to their current location.
Keeping Government Honest: Using Big Data to Fix Big Mistakes
Sometimes, the information data scientists find isn’t what planners or their political bosses might have hoped to see. Seattle’s bike share program, Pronto, for example, recently had to be bailed out by the city council after low ridership numbers and high repositioning rates made it financially untenable for the private partner to maintain. It turned out that most riders were taking the bikes from stations near the tops of hills and riding them down to lower stations; no one was riding them back up again.
But these embarrassing revelations are just part of the learning curve when it comes to data-driven city planning. For a data scientist, they still represent a victory since flaws in public programs have been detected and corrected.
Seattle also provides a good example of this sort of ongoing adjustment. According to a 2013 article in GCN Magazine, a publication covering government information technology use, the city’s power utility, Seattle City Light, has partnered with Microsoft to harvest data on energy use from several large buildings in the downtown core to be analyzed in Microsoft’s Azure cloud computing platform. The accumulated data and subsequent investigation are expected to reveal trends and usage patterns that could reduce power demand in those facilities by up to 25 percent.
Mixing Public and Private Data to Pull Neighborhoods Together
Embedding sensors in pavement to track traffic flow, it turns out, is pretty expensive. New York City spent $100 million in the early 1990s to embed a paltry 1,100 sensors throughout the city. The cost is prohibitive for many lesser-traveled byways or long stretches of highways.
But less than a decade later, most drivers were carrying around their own speed and motion sensor: the average GPS-equipped cellphone. Applications like Waze, from Google, allow users to opt-in to have their speed and position tracked and combined with data from other nearby users to detect traffic flow patterns. Users can also report obstructions or traffic jams manually.
Even in cases where users don’t opt-in to sharing their location information, just the signal strength and direction from phones can be used to provide basic traffic data.
City planners have been quick to jump on the large volume of transportation data these private services accumulate. As described in an article in Forbes, for instance, Florida’s Department of Transportation has signed an agreement to make use of Waze data for tracking and planning.
But the data doesn’t just flow one way. Florida provides Waze with their traffic camera streams and data from their own embedded sensors to improve the information available to Waze users.
All these applications of data science to make the lives of average citizens better and easier are only the tip of the iceberg when it comes to data use in city planning. Each emerged from larger sets of data that enabled those programs in the first place.
In Chicago, for example, data from trackers on each of the plows and salt spreading machines is transmitted back to a city command post and used by traffic planners to orchestrate snow clearing efforts. In Boston, the bike sharing data helps decide where the next station will be placed and how it should be stocked.
As data scientists find new ways to integrate a new wealth of private data sources with the accumulation of public sector data, new insights and benefits will accrue from the combination.