Designers from different disciplines and departments have to consume a lot of time and resources on collaborative actions (calculation, verification, discussion, consensus, etc.) to obtain a compromise decision. As a result, the number of preferred solutions recommended to designers remains large. For complex product design problems with multiple objectives, variables, and constraints, it is indispensable to keep the diversity of solutions so that solutions are sufficiently close to the Pareto front. The Pareto solutions (PS) are unable to realize the enhancement in one optimization objective without deteriorating their performance in at least one of the rest. In recent years, research on the aided design of complex products based on MDO and multi-objective optimization (MOO) has attracted much attention and obtained fruitful achievements, which are applied in various engineering fields such as electromagnetics and space. Further, designers are accustomed to selecting or developing an appropriate product design in detail concerning “preferred” solutions. Therefore, conflicts are always weighed to obtain designer-interested “preferred” solutions, formally known as the Pareto solution set. However, there exists no single optimal solution that optimizes all objectives concurrently. Ĭomplex product design is typically categorized as multi-disciplinary design optimization (MDO) problem, where multi-objective conflicts are widespread. Immense challenges remain in the decision-making of complex product design, stemming from the difficulty of balancing both data and knowledge for mutual benefit. However, due to the proliferation of product complexity and scale, preference determination, with uncertainty and fuzziness, is not only based on data-perceived features but also related to amounts of domain knowledge. Preference-based decision-making with multi-objective conflicts is a widespread and critical task in the preliminary design of complex products. Finally, an engineering application is provided to verify the effectiveness of the proposed method, and the superiority of which is illustrated by comparative analysis. Multi-objective weighted calculations are utilized only to label and compare solutions in the final decision set. Furthermore, fuzzy comprehensive evaluation (FCE) and Pearson correlation are adopted to quantify domain knowledge and obtain abducted labels. In particular, a subtle improvement is presented for WK-means based on the entropy weight method (EWM) to address the local static weight problem caused by the fixed data preferences as the decision set is locally invariant. The proposed G-ABL framework, containing three cores: classifier, abductive kernel, and abductive machine, supports preference integration from data and fuzzy knowledge. This article mainly proposes a novel decision-making method based on generalized abductive learning (G-ABL) to achieve autonomous and efficient decision-making driven by data and knowledge collaboratively. ![]() In recent years, enormous challenges are involved in the design process, within the increasing complexity of preference. However, since complex products involve intensive multi-domain knowledge, preference is not only a comprehensive representation of objective data and subjective knowledge but also characterized by fuzzy and uncertain. ![]() In complex product design, lots of time and resources are consumed to choose a preference-based compromise decision from non-inferior preliminary design models with multi-objective conflicts.
0 Comments
Leave a Reply. |