In complex and critical environments it becomes crucial to execute rapid, adaptive, flexible, predictive, robust decision-making processes and to support collaboration and lateral thinking.
To support these processes, enabling all the desired characteristics, it is necessary to exploit the large amounts of heterogeneous data that emerge from the aforementioned environments.
Hence, the so-called data-driven decision-making comes to life, the complexity of which increases as data increases, in terms of volumes and speed, and as data heterogeneity increases, in terms of type, accuracy, and value offered.
Thus, the study of Big Data was born for which it is also required to act in real-time, that is, to extract useful information and knowledge without having all the data available except for one or more streams of data to process.
The most important technological challenges for data-driven decision-making concern the integration and combination (especially semantics) of data from different information silos to produce value (knowledge), and the selection and collection of "actually" relevant data (for the specific objective). These challenges cannot be solved only with technological systems as there is a need to deploy hybrid human-machine platforms according to the approach called human-in-the-loop. Nevertheless, this approach generates further challenges: men have higher execution times than machines, they suffer from information overload and make mistakes due to incorrect mental models, poor concentration, and other factors.
It is possible to improve the performance of hybrid platforms through the implementation of models for Situation Awareness.
The term Situation Awareness indicates the "awareness" of what is happening around us, understanding it, and being able to project it into the immediate future to make decisions. This is what happens when we want to cross a busy road or when we drive our car. The decisions we make must be immediate and the variables involved to consider are numerous. In this case, the "awareness" of the information and, in particular, of the one that is truly relevant to the current objective, helps us decide on how to act quickly.
Hence, there is a strong need to design and implement human-machine platforms that support and improve Situation Awareness. The application domains are many, for instance, airport security or the management of major events. These platforms must support human operators in perceiving the elements of the monitored environment, recognizing and understanding situations (and anomalous situations), and projecting such situations (to anticipate, for example, possible threats). All this is to support decision-making processes and carry out the right actions in the most appropriate times.
The definition of theories, models, methodologies, and tools capable of supporting the design and implementation of human-machine platforms enabling forms of Situation Awareness is already at the attention of the scientific community.
In this context, the DISA-MIS Research Group on "Models, Methods, and Systems for Situation Awareness" acts which, on the one hand, works on basic and applied research activities (considering different application domains such as Cyber Security, Counterterrorism, Logistics, etc.) with particular reference to Cognitive Architectures, Computational Intelligence and Granular Computing and, on the other hand, guarantees the teaching activities necessary to support the development of skills for the students of the master's degree course in "Business Innovation & Informatics” regarding the design of human-machine systems based on Situation Awareness.