CAP - Computer Aided Principle
CAP (Computer Aided Principle) is a support tool for design principle extraction proposed by Mr. Qiang Yu, Assistant Professor at Faculty of Engineering, Yokohama National University. modeFRONTIER includes specific features for processing CAP in accordance with a contract for technical use of it with Yokohama National University.
HOW IT WORKS
While conventional sensitivity and contribution analyses focus on effects of a single variable, CAP focuses on effects (trends) of combinations of multiple variables. CAP at first classifies data into groups based on characteristic value (clustering), then it visually represents trends in description variables, which characterize each cluster, and variations in the trends in the description variables, which can be seen when moving between clusters. From such visual representations CAP tries to obtain ideas that guide to the physical meaning of these trends, the so called principle.
Therefore with this technique modeFRONTIER is not only able to find what is the optimum, but why it is optimum.