Two-additive fuzzy measure-based information integration approach to product design alternative evaluation

Abstract

Conceptual design is a pivotal stage of new product development in manufacturing industries. Since multiple design alternatives are put forward at this stage, developing advanced evaluation methods is of great importance. Existing methods adopt additive models to integrate evaluation data. They face some inconsistency issues, e.g. inconsistency in the independent assumption and interdependent data, since evaluation criteria are interactional. Fuzzy measure that replaces the additivity with monotonicity has enabled advances in addressing such issues. This work proposes a two-additive fuzzy measure-based information integration approach to product design alternative evaluation for the first time. The evaluation data given by experts are in the form of intuitionistic linguistic numbers. They are more in accordance with the thinking habits of experts because the hesitation degree in linguistic assessment can be revealed. In order to reduce the subjective bias, the decision-making trial and evaluation laboratory method combining with grey relational analysis is applied to adjust evaluation data. Then monotonous two-additive fuzzy measure is identified by nonlinear programming using these data. It makes a good trade-off between computational complexity and presentation capability. Hence, evaluation data can be integrated by non-additive Choquet integral for ranking design alternatives. In comparison to additive model-based methods, the extra effect on the simultaneous satisfaction of criteria can be effectively revealed by the proposed approach. And the robustness of it is demonstrated by the sensitivity analysis. A case study on an elevator's design alternative evaluation is conducted to illustrate the feasibility and practicability of the proposed approach.