In this work, metric vectors for the fair quantitative assessment and comparison of multi-aircraft force compositions with unmanned aerial vehicles (UAV) and/or manned-unmanned teaming (MUM-T) are proposed for specific representative missions in the form of imaging intelligence (IMINT) and close air support (CAS) vignettes. General advantages and disadvantages of force compositions combined with common mission tasks lead to a hierarchically structured pool of possible metrics, which are also known as performance indicators, from which suitable measures are selected for the respective mission type(s). These are tested on data from agent-based constructive simulation. Combining simulation results and the associated expert-derived criteria weights, which represent the importance of the respective items, yields insights about systems effectiveness potentials. Additionally, vignette-specific linear regression, support vector regression (SVR), and neural network regression (NNR) metamodels are derived to enable estimation of mission performance of multi-aircraft force compositions without explicit simulation. These are compared among each other and trialed against test data.
«In this work, metric vectors for the fair quantitative assessment and comparison of multi-aircraft force compositions with unmanned aerial vehicles (UAV) and/or manned-unmanned teaming (MUM-T) are proposed for specific representative missions in the form of imaging intelligence (IMINT) and close air support (CAS) vignettes. General advantages and disadvantages of force compositions combined with common mission tasks lead to a hierarchically structured pool of possible metrics, which are also kno...
»