Chicago advocates open source predictive analytics for city management

A lire sur: http://www.atelier.net/en/trends/articles/chicago-advocates-open-source-predictive-analytics-city-management_426875

By Manon Garnier January 17, 2014 

Chicago is developing SmartData, an open-source program designed to enable the Windy City – and also help other municipal authorities – to take advantage of data flows in order to improve city management.

Each and every day Chicago collects 7 million rows of data on everything from weather to traffic patterns and the location of libraries. However, this abundance of data will not solve urban problems all by itself; being stored in separate, often incompatible, systems, most of the data cannot easily be drawn together. Following a law on Open Data signed by Michael Bloomberg in 2012, New York and a growing number of cities and local organizations are stepping up to exploit these data respecting a principle of open data. Thanks to a $1 million grant from Michael Bloomberg’s private foundationBloomberg Philanthropies, the city of Chicago is developing a toolintended to help collate, sift and analyze all the information available. This tool called SmartData platform is designed to improve local government decision-making by combining relevant data and detecting relationships in the data. Under the ongoing project, millions of lines of data will be analyzed in real time to help make smarter, faster decisions addressing a wide range of urban issues. The aim is that the tool will help to deliver services to Chicago residents more promptly and more closely targeted to their needs.

Towards pro-active city management

The intention is that the SmartData platform will shift city management away from a reactive model towards a proactive approach. The predictive power of the tool resides in its ability to analyze relationships in the data at a speed and on a scale not previously possible, helping Chicago to optimize services of all kinds, benefiting citizens and reducing costs. A pre-cursor to the full-scale SmartData project, a tool called WindyGrid, is already in use. It displays a unified view of operational data via a user-friendly single graphical interface, allowing citizens, security services and other users to make queries and also arrange to receive automatic updates and alerts. WindyGrid incorporates more than a dozen different types of data, including 911 (emergency) calls, public tweets, emergency operations data, city bus location data and video feeds from surveillance cameras. A new pilot project focusing on reducing the rodent nuisance has been launched with a view to demonstrating how Chicago can apply predictive analytics to combat a range of urban issues more effectively. The algorithm used in the pilot identifies and analyzes 31 key call types coming into Chicago’s 311 (non-emergency police calls) call center that serve as predictors of spikes in rodent activity seven days in advance. This will enable the City’s rodent elimination teams to deploy in an earlier, more targeted fashion to prevent serious rodent outbreaks.

Building a replicable, customizable model

The SmartData project has twin goals: firstly to help Chicago city managers analyze trend data and engage in predictive problem-solving; and secondly to share the platform with cities that lack the resources to build a platform for themselves. All software developed on this project will be in open source and made available to other cities, giving them access to the groundwork done in Chicago. They will be able to use the set of predictive algorithms (both logic and source code) and so avoid incurring any startup software development costs. The Chicago Department of Innovation & Technology (DoIT) is also setting up an archive of instructive documents and templates that will provide other cities with a roadmap for developing their own predictive analytics tools. Chief Information Officer Brenna Berman, who is overseeing the SmartData project, stresses that continuing to expand the use of analytics across the city is a top priority. The DoIT team is now developing a new method of data-driven decision-making that looks set to change how municipal operations across the United States are run.