Orienting product performance improvement: application to mechatronics

Abstract

On which criteria design efforts have to be focused on when pursuing the improvement of a product so as to provide most satisfaction to customers? The answer cannot be so obvious at the earliest steps of the design process when available knowledge is still incomplete and vague. Moreover, the problem becomes more complicated when the decision maker has to express some sophisticated preferences on the set of solutions. In this paper we propose to contribute to this issue by proposing an original qualitative computation of the worth-index, the purpose of which is to estimate the mean expected gain in performance improvement on a set of interrelated criteria. An illustration is given in the mechatronics field.

Introducing the difficulty of implementing alternatives in the multiple criteria decision problems

“…Suppose that the student should improve his scores to obtain the scores x i , the difficulty values associated are presented in Table II. Suppose that the student wants to use (5) to assess the difficulty to obtain the score 3 i=1 ω i x i = 12 by improving his first score from x 0 = (8, 10, 12) to x = (12,14,18) in the second test. Considering a Choquet integral for G in (5), we get:…”

“…The following example of µ d (see TABLE III) satisfies the constraints r 1 , r 2 and r 3 . Thus, one can determine the difficulty to obtain the scores (12,14,18) from the scores (8, 10, 12): d x 0 (12, 14, 18) = 6.1 The fact that the student decides to make the greatest improvement in literature makes the global improvement feasible but requiring a high effort.…”