Multicriteria analysis of real-life engineering optimization problems: statement and solution

Abstract

The majority of engineering problems are essentially multicriteria. These criteria are usually contradictory. That is why specialists experience significant difficulties in correctly stating engineering optimization problems, so designers often end up solving ill-posed problems. In general, it is impossible to reduce multicriteria problems to single-criterion ones. For the correct statement and solution of engineering optimization problems, a method called Parameter Space Investigation (PSI method) has been created and widely integrated into various fields of industry, science, and technology (e.g., design of the space shuttle, nuclear reactor, missile, automobile, ship, and metal-tool). In summary, the PSI method generates many feasible designs from which the so-called Pareto optimal ones (i.e. solutions which cannot be improved) are extracted. The PSI method can also be used to efficiently optimize models in a parallel mode, which is of great importance while solving high-dimensional multiparameter and multicriteria problems. The PSI method is implemented in the software package Multicriteria Optimization and Vector Identification (MOVI), a comprehensive system for multicriteria engineering optimization (design, identification, and control). This system allows optimization of many problems that until recently appeared intractable.