Conceptual design decision-making considering multigranularity heterogeneous evaluation semantics with uncertain beliefs

Abstract

Conceptual design decision-making, as an important stage of new product development, is used to select the most appropriate design by decision-makers (DMs) carrying out semantic judgments of conceptual schemes (CSs). However, evaluation semantics involve uncertainties such as subjectivity, randomness, and heterogeneity, and how to capture and handle these uncertainties is one of the challenging topics. Previous studies have focused on the introduction of heterogeneous semantics with different expressions for CS ranking while neglecting the negative impact of multigranularity semantics as well as fuzzy cognition on CS decision results. Furthermore, the randomness embedded in the evaluation semantics has not yet been addressed in heterogeneous semantics integration. To fill these gaps, a conceptual design decision-making approach considering multigranularity heterogeneous evaluation semantics with uncertain beliefs is proposed. First, multigranularity semantics with uncertain beliefs are assigned by DMs, and an initial decision matrix for CSs with a mixed distribution of crisp values, interval values, and monotonic values is constructed. Second, two transformation rules for the normal cloud model and the exponential cloud model are constructed to transform the multigranularity heterogeneous semantics into cloud models, and the cloud model decision matrix is built. Third, by generating random numbers based on the distribution of cloud models to construct the dominance matrix, the Vlsekriterijumska Optimizacija I Kompromisno Resenje (VIKOR) model is used to select the optimal CS. A case study of a tree climbing and trimming (TCT) machine is used to validate the effectiveness of the proposed model, and the sensitivity analysis and comparison results confirm the practicality of the concept selection process considering the multigranularity heterogeneous evaluation semantics.