Hierarchical and Interactive Decision Refinement Methodology for Engineering Design

摘要: This paper presents the Hierarchical and Interactive Decision Refinement (HIDER) methodology for engineering design. hierarchically refines (or reduces) a large initial design space through series of multiple-objective optimizations, until fully specified is obtained. contrasts with traditional approach to optimization which involves picking from possible designs then iteratively modifying its specifications maximize performance. The HIDER uses an adaptive modeling that combines machine learning statistical techniques developing fast empirical models different levels detail evaluation speeds. These are used at stages use layered enables be parametric design, including early when not complete. also supports multiple perspectives in concurrent because it provides means transforming these into uniform representation, as well facilitating decision-making respect competing objectives. example diesel engine.