A model of mission accomplishment in simulated battle
Gary S. Thomas, Thomas G. Cocklin, Systems Research Laboratory (U.S. Army Research Institute for the Behavioral and Social Sciences)
U.S. Army Research Institute for the Behavioral and Social Sciences, 1983 - Business & Economics - 19 pages
"This research sought to formulate a composite measure of mission accomplishment that could be used to assess simulated combat performance of battalion command groups, training command and control procedures, in a computer-driven battle simulation. Mission accomplishment was defined in terms of six measurable components or objectives of a covering force/delay mission. Four retired Army officers served as judges for all experiments. Experiment I demonstrated that two of the objectives could be meaningfully combined into a single measure that was used in previous research as an index of relative losses by opposing forces. Multiple regression was used in Experiment II to describe how the judges assigned mission accomplishment scores to 216 hypothetical battle outcomes including measures of each mission objective. Experiment III compared judges' assessments of 10 actual battle outcomes to the mission accomplishment scores predicted by the regression models derived in Experiment II. Inter-rater agreement was high in the assessments of actual battle outcomes, and individual predictor models accounted for more than 94% of the variance in ratings of these data for three of four judges. A composite model of mission accomplishment accounted for 93% of the variance in the average assessments of all four judges. Results indicate that it may be feasible to replace the expert judges with the composite model of mission accomplishment. Originator supplied keywords include: Performance prediction, Performance evaluation, Policy capturing, and Decision making."--Stinet.
What people are saying - Write a review
We haven't found any reviews in the usual places.
ACCOMPLISHMENT IN SIMULATED actual battle outcomes actual CATTS data actual CATTS exercises actual mission accomplishment Army Research Institute assigning mission accomplishment attack data average mission accomplishment battalion command groups battle simulations battlefield performance BCG performance Behavioral and Social Combined Arms command and control composite measure composite model conclusion of battle correlations covering force data covering force mission decision rules Experiment force mission objectives four judges friendly forces surviving Hirsch hypothetical battle outcomes individual models individual regression models INTEL inter-rater agreement judges combined Kaplan linear regression loss data main effects measure of mission military judges mission accomplishment ratings mission accomplishment scores mission type model of mission multiple regression Naylor NUMBER OPFOR surviving opposing forces percent of friendly percentage of friendly Percentage of OPFOR predicted average mission predicted judges predicted mission accomplishment Procedures raters regression equations relative losses simulated battle Slovic SMFRD Thomas U.S. Army Research UNCLASSIFIED variance accounted variance in mission