DEVELOPMENT OF AN EXPERT-SYSTEM FOR SELECTION OF EXPERIMENTAL-DESIGNS

An expert system has been developed to assist chemists in the selectio n of experimental designs for research projects. The system (named DXP ERT) ranks thirteen types of experimental designs according to their s uitability for projects presented by users in an interactive session. Design categories included are factorial, response surface, sequential simplex optimization, simplex mixture, and statistical testing. A des irability index is assigned to each design alternative according to pr oject characteristics (attributes). Characteristics are interpreted ba sed on expert knowledge built into the system. DXPERT uses mathematica l concepts to mimic features of human intuition and decision making. E xpert knowledge is represented by relevance factors (a number between minus one and plus one) in a multiple-alternative multiple-attribute t able. Relevance factors are interpreted as fuzzy values that represent the degree to which a design belongs to the set of suitable designs. The formula for calculating design desirabilities is based on fuzzy ma thematics. For efficiency purposes, the order of the questions present ed to the user is driven by a maximum potential information gain algor ithm. Design desirability indexes were found to be useful to researche rs for the elimination of unsuitable designs and concentration of furt her efforts in the most applicable designs. A validation test was cond ucted with the participation of four other experts in the field.