The presentation introduced a basic proactive model for taking timely actions to optimize outcomes given anticipated unplanned events. It uses event patterns to predict events and their timing with uncertainty, then determines the optimal action over time by considering action costs and impacts. Several application scenarios were discussed. While demonstrating feasibility, further work is needed to address challenges like real-time optimization for other cases, improved forecasting models, usability, and scalability.
Proactive event driven applications follow a 4 stage pattern: Detect phase – we monitor and detect interesting events. In our case we get the location, time, and magnitude of the earthquake from seismic sensors and damages from reports of citizens (uploaded questionnaires on-line to the web). Forecast phase – based on the detected events and causality models we calculate the potential loss and deformation Decide phase – based on the forecasted events we decide in real-time the steps and protocols to be followed Act phase – upon on the decision some actions are taken. Note that these can be automatic, e.g. broadcasting alerts, stopping a train, closing a bridge, closing a nuclear plant, or recommended actions like send troops or equipment to certain area, closing of places and evacuation of people