Chama Research
Optimization Approaches Scenario Vs Time Scenario
Convert detection times to input for the Coverage Formulation The Coverage Formulation requires as input a DataFrame with two columns: ‘Sensor’, and ‘Coverage’, where the ‘Coverage’ is a list of entities covered by each sensor. The formulation optimizes a sensor layout that maximizes the coverage of the entities contained in this DataFrame. An entity to be covered might include scenarios, scenario-time pairs, or geographic locations.
Scenario
Scenario turns detection times into coverage. scenario_cov = chama.impact.detection_times_to_coverage(det_times, coverage_type='scenario') We set the coverage_type to scenario.
Scenario-Time
The next example converts detection times to scenario-time coverage. With scenario-time coverage, the entities to be covered are all combinations of the scenarios and the detection times. This type of coverage gives more weight to sensors that detect scenarios for longer periods of time. The same detection_times_to_coverage function can be used to convert detection times to scenario-time coverage with one major difference to the previous case. With scenario coverage the scenarios themselves become the entities to be covered. This means that if there is additional data available for the scenarios such as weights/probabilities or undetected impact, these values can be used directly in the coverage solver. With scenario-time coverage, we are essentially defining new entities/scenarios. So any data corresponding to the original scenarios must be translated to the new entities before they can be passed to the coverage solver. The detection_times_to_coverage function does this by accepting an optional ‘scenario’ keyword argument containing a DataFrame with scenario probabilities and undetected impact. These values are then propagated to the new scenario-time entities and a new DataFrame is returned with this information.
Convert detection times to scenario-time coverage and propagate scenario information to new scenario-time pairs: