Abstract
The purpose of this paper is to present the components of an Artificial Intelligence (AI)-based system design for better city strategic risk control. Several smart cities have made open data available to support various stakeholders’ interests: web pages and data tables can guide citizens and businesses and in many cases enable them to carry out service transactions. The data provides a resource for city studies, the development of indicators in support of city policymakers, and city administrators. However, the problem of administering a city requires scaling up to an integrated view, so such
systems are at an early stage of development. This study presents example cases where a dynamic and predictive system for a city has been created based on the use of AI, to guide city administrators based on possible future events. The cases cover crime, road traffic management/accidents, education, and health events, using data from three North American cities: Baltimore, Chicago, and Toronto. Together the cases serve as both a proof of concept for, and a test of, the approach needed to create an integrated predictive system. In this paper, the AI models are described along with all the steps in the approach, from data gathering to the creation of a system to support decisions. The main points are related to how risks can be mitigated and controlled using AI in strategy and policy formulation and implementation to improve citizens’ life. Data patterns can drive decisions, such as: crime seasonality supporting the planning of patrols and human presence in areas of potential issues; understanding traffic levels reducing the time people spend in cars; coordination of investment
providing a better use of the city’s resources. The examples presented illustrate the creation of a range of dynamic and adaptive predictive systems based on AI that are fed by the city-generated open data which contributes to the control of
the services provided by the city. Together they illustrate the feasibility of progress towards fully integrated systems.
systems are at an early stage of development. This study presents example cases where a dynamic and predictive system for a city has been created based on the use of AI, to guide city administrators based on possible future events. The cases cover crime, road traffic management/accidents, education, and health events, using data from three North American cities: Baltimore, Chicago, and Toronto. Together the cases serve as both a proof of concept for, and a test of, the approach needed to create an integrated predictive system. In this paper, the AI models are described along with all the steps in the approach, from data gathering to the creation of a system to support decisions. The main points are related to how risks can be mitigated and controlled using AI in strategy and policy formulation and implementation to improve citizens’ life. Data patterns can drive decisions, such as: crime seasonality supporting the planning of patrols and human presence in areas of potential issues; understanding traffic levels reducing the time people spend in cars; coordination of investment
providing a better use of the city’s resources. The examples presented illustrate the creation of a range of dynamic and adaptive predictive systems based on AI that are fed by the city-generated open data which contributes to the control of
the services provided by the city. Together they illustrate the feasibility of progress towards fully integrated systems.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the European Conference on the Impact of Artificial Intelligence and Robotics ECIAIR 2019 |
| Editors | Paul Griffiths, Mitt Nowshade Kabir |
| Place of Publication | Reading, UK |
| Publisher | Academic Conferences and Publishing International |
| Pages | 277-286 |
| ISBN (Electronic) | 978-1-912764-44-0 |
| ISBN (Print) | 978-1-912764-45-7 |
| Publication status | Published - 31 Oct 2019 |
| Event | European Conference on the Impact of Artificial Intelligence and Robotics - Oxford, United Kingdom Duration: 31 Oct 2019 → 1 Nov 2019 |
Conference
| Conference | European Conference on the Impact of Artificial Intelligence and Robotics |
|---|---|
| Country/Territory | United Kingdom |
| City | Oxford |
| Period | 31/10/19 → 1/11/19 |
Keywords
- artificial Intelligence
- smart cities
- strategic risk
- analytics
- dynamic and predictive performance systems