An experiment to evaluate the effectiveness of JUDGE (Judged Utility Decision Generator), an advanced man-machine technique for allocating and dispatching missions in response to requests for immediate close air support when resources are limited. The subjects were 14 junior and senior Army ROTC cadets from UCLA. They worked 2 hours a day for 8 weeks and were given extensive training, longer scenarios than in the earlier field study (see RM-5147-PR), and repeated measurements. For comparison, JUDGE was pitted against another technique called DASC (for Direct Air Support Center), a hypothetical version of the current Air Force system. Using JUDGE, each subject evaluated the target worth by comparing it with a carefully defined standard target (taken as 100). A computer program then translated these responses into dispatching decisions and evaluated them based on the subject's value responses. Using DASC, the subject himself assigned aircraft to target, allocating any even number from 0 to 16. JUDGE is shown to be superior to DASC when measured by an expected utility criterion. JUDGE performed at a 90 percent level, and DASC at only 40 percent. JUDGE apparently gains part of its advantage by turning over the mechanical computations to the computer.
Miller, Louis W., Richard J. Kaplan, and Ward Dennis Edwards, JUDGE: A Laboratory Evaluation. Santa Monica, CA: RAND Corporation, 1968. https://www.rand.org/pubs/research_memoranda/RM5547.html. Also available in print form.
Miller, Louis W., Richard J. Kaplan, and Ward Dennis Edwards, JUDGE: A Laboratory Evaluation, Santa Monica, Calif.: RAND Corporation, RM-5547-PR, 1968. As of October 26, 2021: https://www.rand.org/pubs/research_memoranda/RM5547.html