The United States, as the world's largest producer and consumer of scrap and flake mica, has an obvious economic interest in applications of artificial intelligence technology that would expedite beneficiation of mica. In the fall of 1985, personnel of the Tuscaloosa and Rolla research centers of the Bureau of Mines and of the University of Missouri-Rolla, Institute for Artificial Intelligence started the following long term research project: Develop an Expert System Consultant for the three basic stages of mica beneficiation: Stage 1: characterization of the material; Stage 2: treatment to obtain a concentrate; and Stage 3: evaluation of the resulting concentrate. Completion of the long term project will result in an Expert System Consultant (MICA) that will assist in providing the following advice for an ore body being considered for development: (1) usability of the non-mica present; (2) feasible end uses for the mica present; (3) concentration treatments for mica; (4) equipment configurations for the concentration treatments; (5) setting the parameters of the treatment circuits; and (6) fine tuning of the circuits. The consultant will identify the laboratory treatments to be investigated to determine best plant scale flowsheets that will result in commercial operations. To date, a prototype of Stage 1 and a detailed introduction to the consultant have been developed. Stage I provides the basis for recommendations concerning the usability of non-mica, the feasible end uses for mica, and the concentration treatments for mica. This paper describes and discusses the development of the project and the results achieved.


Computer Science


This report is substantially the M.S. thesis of the first author, completed July 1986.

Report Number


Document Type

Technical Report

Document Version

Final Version

File Type





© 1986 University of Missouri--Rolla, All rights reserved.

Publication Date

01 Jul 1986