SOME PRINCIPLES FOR THE DESIGN OF DECISION SYSTEMS: A REVIEW OF SIX YEARS OF RESEARCH ON A COMMAND-CONTROL SYSTEM SIMULATION.
Abstract
A set of general principles for guidance in decision system development is presented based upon research findings obtained in a simulated (but highly generalized) command-control system. The chief objective of the research was evaluation of an automated procedure for assisting man in making diagnostic decisions. Briefly, this procedure involved aggregation by the machine of human evaluations of a number of separate items of reconnaissance data. Comparisons were made between performance of the system in assessing environmental states (e.g., enemy strategies) when the automated procedure was and was not in effect under a variety of task circumstances (e.g., load, level of information fidelity, feedback, etc.). The 13 principles thus formulated generally support the use of an automated aggregation procedure in diagnosis. Furthermore, they show that machine aggregation is most beneficial in circumstances which produce large amounts of low-quality data or those in which the human is placed under some sort of stress. Several of the principles suggest possible limitations on the use of machines in decision making. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1967
- Accession Number
- AD0665469
Entities
People
- William C. Howell
Organizations
- Ohio State University