Fuzzy stochastic multiobjective programming (Q625594)

This is an excellent textbook on fuzzy stochastic multiobjective programming. In this book the authors introduce the latest advances in the new field of multiobjective programming involving fuzziness and randomness. The book contains 6 chapters. Each chapter, except the first, includes numerical examples. A selective bibliography follows the chapters. The chapter headings are as follows: 1. Introduction and historical remarks; 2. Fundamentals; 3. Fuzzy multiobjective stochastic programming; 4. Multiobjective fuzzy random programming; 5. Stochastic and fuzzy random two-level programming; 6. Future research directions. The first chapter has an introductory character and presents some historical remarks of the main concepts of stochastic and fuzzy multiobjective programming. It also summarizes the contents of each chapter. Chapter 2 discusses the mathematical background, especially from fuzzy set theory, linear programming problems with fuzzy goals and fuzzy constraints and linear programming problems with fuzzy parameters. Also, this chapter discusses a number of topics from stochastic programming such as two-stage programming and chance constrained programming. The solution concepts in multiobjective programming and interactive fuzzy multiobjective programming are presented. Several basic notions including Pareto optimal solutions and Stackelberg equilibrium in two-level programming under cooperative and noncooperative decision making situations are reviewed. Fundamentals of genetic algorithms are also briefly discussed as efficient meta-heuristics for solving difficult classes of optimization problems with nonconvex nonlinear objective functions and/or constraints. Chapter 3 presents fuzzy multiobjective stochastic programming. First, the main ideas of conventional single-objective stochastic programming problems such as two-stage problems and chance constrained programming problems, multiobjective stochastic programming problems are formulated by assuming that the decision maker (DM) has a fuzzy goal for each of the expectation and the variance of the original stochastic objective functions. The authors present different optimization criteria from the expectation and the variance, i.e., the probability model maximizing the probability of the objective functions being greater than or equal to target values and the fractile model optimizing the target values under a given probability. These problems are transformed into deterministic multiobjective problems with linear, linear fractional or nonlinear objective functions and/or constraints. Interactive fuzzy satisficing methods using linear or convex programming techniques are presented to derive a satisficing solution for the DM from among the extended Pareto optimal solution set. At the end of this chapter, these interactive methods for the multiobjective stochastic programming with continuous variables are extended to deal with integer problems for which the authors employ genetic algorithms to derive a satisficing solution to the DM. Chapter 4 discusses multiobjective programming problems with fuzzy random variables, taking into account not only the randomness of parameters involved in objective functions and/or constraints but also the experts' ambiguous understanding of the realized values of the random parameters. Four types of optimization models for fuzzy random programming are developed by incorporating a concept of possibility measure into previous stochastic programming models. The authors introduce an extension concept of Pareto optimal solutions on the basis of possibility theory and probability theory and show the development of interactive methods for fuzzy random multiobjective programming. Some Pareto optimality concepts for fuzzy random multiobjective programming problems are defined by combining the notions of the \(M\)-\(\alpha \)-Pareto optimality and optimization criteria in stochastic programming. Interactive satisficing methods for deriving a satisficing solution of the DM from the extended \(M\)-\( \alpha \)-Pareto optimal solution are presented. The presented interactive methods are extended to deal with integer programming problems with fuzzy random variables. Chapter 5 discusses two-level programming problems with random variable and fuzzy random variable coefficients. By employing the optimization models from Chapter 3 and 4, the original problems are transformed into deterministic two-level programming problems for which interactive fuzzy programming methods for cooperative stochastic and fuzzy random two-level programming problems are developed. Also, computational methods using the combination of convex programming techniques and the branch-and-bound method are presented in order to obtain Stackelberg solutions to noncooperative methods and fuzzy random two-level programming problems. The presented methods are extended to deal with two-level integer programming problems with random variables and fuzzy random variables, using genetic algorithms. Chapter 6 discusses a number of topics from which further research and application developments are expected. Now, random fuzzy variables are introduced as parameters or coefficients involved in linear programming problems by considering the ambiguity of means and variances of random variables. The case where the mean of each of random variables is represented by a fuzzy number is considered. The original random fuzzy linear programming problem is equivalently transformed into a deterministic nonlinear programming problem. The model is extended to deal with multiobjective and two-level programming problems with random fuzzy variables. The book is complemented by an index. The book is beautifully written and inspiring as a source for further research. The style is clear and rigorous. I warmly recommend this book and also other publications of these authors. Everyone who is interested in optimization should be acquainted with this work.