Objectives
Artificial Intelligence tools to govern soft targets in pandemic scenarios
The main goal of OPENNESS is to design a system based on artificial intelligence that determines the optimal behavioural logic limiting the spread of an epidemic disease in critical infrastructures.
The target of OPENNESS is the adoption of innovative technologies to reduce the spread of an epidemic disease, such as the current COVID-19 pandemic. Achieving the main and intermediate results of OPENNESS will allow citizens to remain active, continuing to use critical infrastructures such as airports, railway stations, and museums while maintaining individual security.
Scientific and Technological Methodology
In the context of COVID-19 pandemic, OPENNESS aims at providing a tool for effectively reducing the spread rate of COVID-19, and similar infection diseases, by supporting the decision making process of those who are in charge of defining strategies to prevent from further infections. By analysing the dynamic of the SARS-CoV-2 virus transmission, we will investigate and model transmission patterns in a variety of different environments.
The key component of our approach is an environment simulator integrated with closed loop mechanisms that make use of the real data captured from the environment. The simulator outputs a forecast that predicts the effects of the SARS-CoV-2 virus spread in response to specific behaviours adopted in “soft target” environments, such as squares, theatres, subway stations, and commuting zones of urban areas.
By using the quantitative data obtained from the simulation, OPENNESS will make use of reinforcement learning to elicit the logic behind the observed behaviours as a set of rules. These rules will be specified within a mathematical framework to verify their correctness and validate their use within the decision making process. This verification and validation process will be conducted by using well established techniques from software engineering.
Overall Strategy
The objectives of OPENNESS are pursued thanks to the synergy of the skills of the Institute of Systems Analysis and Computer Science "Antonio Ruberti" of the National Research Council of Italy and the Department of General and Applied Hygiene of the Catholic University of the Sacred Heart.
The skills in the field of public health and epidemiology of the Department of General and Applied Hygiene of the Catholic University of the Sacred Heart are used to evaluate, on the basis of the available evidence, the diffusion index of the SARS-CoV-2 based on different environmental situations.
The skills in the fields of reinforcement learning, engineering methods, and software systems verification of the Institute of Systems Analysis and Computer Science "Antonio Ruberti" of the National Research Council of Italy, instead, allow to create a realistic virtual environment that is able to effectively determine how different behavioural logic affect the spread of an epidemic, to use this simulator to find the behavioural logic that minimises the risk of spreading the epidemic, and to determine the degree of robustness of these behavioural logic.
