R.B. Statnikov

method enables readers to: *efficiently design higher-quality, lower cost objects with less metal requirements, vibration and noise, and with lower dynamic loads and energy consumption *determine optimal solutions, regardless of the number of criteria involved, and to identify relationships among different criteria and between criteria and design variables *accurately account for discrepancies between theoretical and actual characteristics, using a special set of adequacy criteria *determine optimal design variables for complex finite element models In addition, the book helps readers: *enhance the potential of the PSI method with theoretical investigations and algorithms for approximating the feasible solutions set and Pareto optimal set *facilitate proficient problem-solving by incorporating recently obtained results from the theory of uniformly distributed sequences *evaluate design procedures by observing examples ranging from machine tools and agricultural equipment to automobiles and aviation This practical, in-depth treatment of multicriteria optimization and engineering is essential for engineers and designers working in research and development of manufacturing machines, mechanisms and structures. It is also an important text for students of applied mathematics, mechanical engineering, optimal control and operations research.

This book is devoted to the PSI method. Its appearance was a reaction to the unsatisfactory situation in applications of optimization methods in engineering. After comprehensive testing of the PSI method in various fields of machine engineering it has become obvious that this method substantially surpasses all other available techniques in many respects. It has now become known that the PSI method is successfully used not only in machine design, at which it was initially aimed, but also in polymer chemistry, pharmacy, nuclear energy, biology, geophysics, and many other fields of human activity. To all appearances this method has become so popular for its potential of taking into account the specific features of applied optimization better than other methods, being, at the same time, comparatively simple and friendly, and because, unlike traditional optimization methods which are intended only for searching for optimal solutions, the PSI method is also aimed at correctly formulating engineering optimization problems. One well-known aircraft designer once said, "To solve an optimization problem in engineering means, first of all, to be able to state this problem properly". In this sense the PSI method has no competitors. Although this method has been presented in Russia in numerous papers and books, Western readers have had the opportunity to familiarize themselves with this method only recently (Ozernoy 1988; Lieberman 1991; Stadler and Dauer 1992; Dyer, Fishburn, Steuer, Wallenius, and Zionts 1992; Steuer and Sun 1995, etc. ).

Optimization methods have been considered in many articles, monographs, and handbooks. However, experts continue to experience difficulties in correctly stating optimization problems in engineering. These troubles typically emerge when trying to define the set of feasible solutions, i.e. the constraints imposed on the design variables, functional relationships, and criteria. The Parameter Space Investigation (PSI) method was developed specifically for the correct statement and solution of engineering optimization problems. It is implemented in the MOVI 1.0 software package, a tutorial version of which is included in this book. The PSI method and MOVI 1.0 software package have a wide range of applications. The PSI method can be successfully used for the statement and solution of the following multicriteria problems: design, identification, design with control, the optional development of prototypes, finite element models, and the decomposition and aggregation of large-scale systems. Audience: The PSI method will be of interest to researchers, graduate students, and engineers who work in engineering, mathematical modelling and industrial mathematics, and in computer and information science.

Optimization methods have been considered in many articles, monographs, and handbooks. However, experts continue to experience difficulties in correctly stating optimization problems in engineering. These troubles typically emerge when trying to define the set of feasible solutions, i.e. the constraints imposed on the design variables, functional relationships, and criteria. The Parameter Space Investigation (PSI) method was developed specifically for the correct statement and solution of engineering optimization problems. It is implemented in the MOVI 1.0 software package, a tutorial version of which is included in this book. The PSI method and MOVI 1.0 software package have a wide range of applications. The PSI method can be successfully used for the statement and solution of the following multicriteria problems: design, identification, design with control, the optional development of prototypes, finite element models, and the decomposition and aggregation of large-scale systems. Audience: The PSI method will be of interest to researchers, graduate students, and engineers who work in engineering, mathematical modelling and industrial mathematics, and in computer and information science.

This book is devoted to the PSI method. Its appearance was a reaction to the unsatisfactory situation in applications of optimization methods in engineering. After comprehensive testing of the PSI method in various fields of machine engineering it has become obvious that this method substantially surpasses all other available techniques in many respects. It has now become known that the PSI method is successfully used not only in machine design, at which it was initially aimed, but also in polymer chemistry, pharmacy, nuclear energy, biology, geophysics, and many other fields of human activity. To all appearances this method has become so popular for its potential of taking into account the specific features of applied optimization better than other methods, being, at the same time, comparatively simple and friendly, and because, unlike traditional optimization methods which are intended only for searching for optimal solutions, the PSI method is also aimed at correctly formulating engineering optimization problems. One well-known aircraft designer once said, "To solve an optimization problem in engineering means, first of all, to be able to state this problem properly". In this sense the PSI method has no competitors. Although this method has been presented in Russia in numerous papers and books, Western readers have had the opportunity to familiarize themselves with this method only recently (Ozernoy 1988; Lieberman 1991; Stadler and Dauer 1992; Dyer, Fishburn, Steuer, Wallenius, and Zionts 1992; Steuer and Sun 1995, etc. ).

method enables readers to: *efficiently design higher-quality, lower cost objects with less metal requirements, vibration and noise, and with lower dynamic loads and energy consumption *determine optimal solutions, regardless of the number of criteria involved, and to identify relationships among different criteria and between criteria and design variables *accurately account for discrepancies between theoretical and actual characteristics, using a special set of adequacy criteria *determine optimal design variables for complex finite element models In addition, the book helps readers: *enhance the potential of the PSI method with theoretical investigations and algorithms for approximating the feasible solutions set and Pareto optimal set *facilitate proficient problem-solving by incorporating recently obtained results from the theory of uniformly distributed sequences *evaluate design procedures by observing examples ranging from machine tools and agricultural equipment to automobiles and aviation This practical, in-depth treatment of multicriteria optimization and engineering is essential for engineers and designers working in research and development of manufacturing machines, mechanisms and structures. It is also an important text for students of applied mathematics, mechanical engineering, optimal control and operations research.