About COVID-19 and AI Simulation Project

　The COVID-19 AI Simulation Project explores the potential use of technology to balance economic activities with measures to prevent the spread of COVID-19.
We collect and analyze data on the early detection of the spread of infection using AI and other technologies, conduct simulations, develop new technologies that contribute to infection prevention measures, and verify the results for social implementation.

Approaches from multiple perspectives are necessary to predict and respond to uncertain events such as the recent COVID-19 pandemic. It is important to discuss and examine the validity of various models and verification methods, and to encourage further refinement by exchanging opinions among teams and models adopting different approaches.
For example, in predicting infection, it is common to quantitatively grasp the infection situation using an infectious disease mathematical model called SEIR. On the other hand, in the actual world, various infection types can occur according to complex human movements. Thus, we consider the multi-agent model and complex network theory also useful, dynamically linking various elements in analyzing how infection spreads.
In addition, it is important as a basic scientific attitude to update hypotheses and simulation assumptions in accordance with the latest data and situations that are constantly updated. In this project, discussions are held as needed among teams that adopt multiple approaches, and the latest results are announced promptly based on analyses that are tailored to the strengths and characteristics of each model.

　In this project, instead of coming to a simple prediction, we analyze various scenarios such as “What could happen if … takes place.” In situations where the prediction itself causes people’s behavioral changes, it is important to consider multiple policies that accommodate various possibilities under multiple scenarios, rather than providing a single-scenario solution to the uncertain situation.
　The spread of COVID-19 is affected by a complex interplay of uncertain and diverse factors, such as the progress of vaccination, the amount of medical resources available, and the effects of declaring an emergency. The purpose of this project is to present materials for policy making assuming various scenarios using multiple analytical models.

　There is a need for a system that can quickly provide scientific evidence in response to ever-changing situations. In this project, individual R & D teams and Open Collaboration Partners (OCP) (hereinafter referred to as “Research Teams”) have been selected by open recruitment for each R & D area. In addition, from the viewpoint of promptly disclosing the research results and ensuring scientific validity, each expert committee member gives advice from a professional point of view for each individual R & D theme, and the progress of the research is monitored as needed. It is announced.

　Various measures are being taken to put an end to the novel coronavirus; however, the fight against the pandemic will not end with the novel coronavirus. To prepare for other pandemics in the future, it is important to accumulate and disclose analytical knowledge, leaving its traces in history.
　We know that real-time access to a variety of data is necessary to accurately analyze infection conditions and make meaningful simulations. Data on highly granular human flow density, human flow movements, genome sequences, international immigration status, cluster analysis, and medical resources are some of these examples. Through this project, we work on the construction of system to access such data, and organize the knowledge that contributes to discussions on the ideal pandemic countermeasures for national security purposes.