Showing papers on "Multi-objective optimization published in 1984"
TL;DR: In this paper, the authors studied the application of multiobjective optimization techniques to structural design problems involving uncertain parameters and random processes and found that the game theory approach is superior in finding a better optimum solution, assuming the proper balance of the various objective functions.
Abstract: The application of multiobjective optimization techniques to structural design problems involving uncertain parameters and random processes is studied. The design of a cantilever beam with a tip mass subjected to a stochastic base excitation is considered for illustration. Several of the problem parameters are assumed to be random variables and the structural mass, fatigue damage, and negative of natural frequency of vibration are considered for minimization. The solution of this three-criteria design problem is found by using global criterion, utility function, game theory, goal programming, goal attainment, bounded objective function, and lexicographic methods. It is observed that the game theory approach is superior in finding a better optimum solution, assuming the proper balance of the various objective functions. The procedures used in the present investigation are expected to be useful in the design of general dynamic systems involving uncertain parameters, stochastic process, and multiple objectives.
89 citations
TL;DR: In this paper, a multiobjective optimization procedure using the constraint linear programming technique is developed to assist water resources planners to establish a more unified basin-wide management plan which will simultaneously consider the three major aspects in basinwide planning: (1) Water supply allocation; (2) water quality control; and (3) prevention of undesirable overdraft of the ground-water basin.
Abstract: A multiobjective optimization procedure using the constraint linear programming technique is developed to assist water resources planners to establish a more unified basin-wide management plan which will simultaneously consider the three major aspects in basin-wide planning: (1) Water supply allocation; (2) water quality control; and (3) prevention of undesirable overdraft of the ground-water basin. This optimization procedure is designed to be applied in concert with one or more simulation models: A ground-water quantity-and-quality model or a river flow-and-mass transport model, or both, and a model which routes the water from the supply sources to various demand points, through the wastewater collection and treatment facilities, and ultimately to the disposal sites. To create the necessary linkage between the optimization procedure an the simulation models, the influence coefficient method is used as the connecting agent. With this agent, the optimization procedure can be carried out independently of the simulation models, while, at the same time, keeping a good representation of the response surface of the simulation models in the optimization procedure.
61 citations
TL;DR: Three approaches to solving design optimization problems with multiple conflicting objectives are described and compared by considering their computational efficiencies and their abilities to produce an approximation of the Pareto optimal set.
Abstract: This paper describes and compares three approaches to solving design optimization problems with multiple conflicting objectives. The three techniques are described in detail and then applied lo an example which demonstrates how information is accumulated which leads to a logical and efficient multicriteria optimal design. The techniques employed (weighting, noninferior set estimation and constraint methods) ate compared to each other by considering their computational efficiencies and their abilities to produce an approximation of the Pareto optimal set.
58 citations
TL;DR: In this paper, an optimization model is presented which is suited to energy planning on the regional level within a free-market economy, where a set of linear constraints determines which combinations of energy conversion processes are technologically feasible.
45 citations
TL;DR: In this paper, the authors have proposed a method called the satisficing trade-off method in the middle of optimization and satisficing, which is more attractive as the decision rule in developing interactive multiobjective programming.
24 citations
TL;DR: In this paper, a multiobjective design optimization method was presented to evaluate conflicting objectives in designing machine-tool spindles, and a Pareto optimum set which showed a trade-off relationship between two objectives minimizing the total weight of the spindle and the static torsional or bending compliance was derived using the Kuhn-Tucker necessary conditions for optimality.
Abstract: A method of multiobjective design optimization is presented in order to evaluate conflicting objectives in designing machine-tool spindles. A Pareto optimum set which shows a trade-off relationship between two objectives minimizing the total weight of the spindle and the static torsional or bending compliance, is derived using the Kuhn-Tucker necessary conditions for optimality, and through analyses of objective and constraint functions. Weighting factors between the two objectives are obtained on the Pareto optimum set. Expanded problems and numerical examples are given.
24 citations
Book•
01 Jan 1984
TL;DR: This work discusses the development of an Interactive Method Supporting Collective Decision Making Using a Regional Development Model and applications of this method to Aid Budget Allocation and Investment and Financial Planning in a General Partnership.
Abstract: I. Introductory Lecture.- Interactive Decision Analysis and Interpretative Computer Intelligence.- II. Approaches and Concepts in Interactive Decision Analysis.- DIDASS - Theory, Implementation and Experiences.- Handling Conflicts in Fuzzy Multiple-Criteria Optimization.- A Game-Theoretic Approach for Multicriteria Decision Making.- The Systems Approach and Contingency View in Managerial Behavior and Management Organization.- Recent Developments in our Approach to Multiple-Criteria Decision Making.- Interactive Structure Design and Simulation of Nonlinear Systems from a Multiobjective Viewpoint Using the Lotka-Volterra Approach.- Hierarchical Model-Oriented System Organization.- III. Methods and Techniques for Interactive Decision Analysis.- Dual Relaxation and Branch-and-Bound Techniques for Multiobjective Optimization.- Levitin-Miljutin-Osmolovskii Conditions for Local Pareto Optimality.- Fuzzy Assessment of Multiattribute Utility Functions.- Interactive Fuzzy Decision Making for Multiobjective Nonlinear Programming Problems.- Satisficing Trade-Off Method for Multiobjective Programming.- On Dialogue Algorithms for Linear and Nonlinear Vector Optimization from the Point of View of Parametric Programming.- Operating Considerations Pertaining to the Interactive Weighted Tchebycheff Procedure.- A Posteriori Trade-Off Analysis in Reference Point Approaches.- A Visual Interactive Method for Solving the Multiple-Criteria Problem.- On the Implementation of the Interactive Surrogate Worth Trade-Off (ISWT) Method.- IV. Applications of Interactive Decision Analysis.- Multicriteria Decision Analysis to Aid Budget Allocation.- Investment and Financial Planning in a General Partnership.- A Multiobjective Expert System for Suppliers of Out-of-the-Money Options.- A Multiple-Layer Model for Economic-Environmental-Energy Policy Analysis.- Simulation of an Interactive Method Supporting Collective Decision Making Using a Regional Development Model.- A Multiobjective Observation Network Design Procedure and its Application in Hydrology and Mining.- Analysis of Regional Water Policies in Open-Cast Mining Areas - A Multicriteria Approach.- Decision Support via Simulation for a Multipurpose Hydroenergetic System.- Multiobjective Analysis of Forestry Management Models Using the Generalized Reachable Set Method.- A Multiobjective Procedure for Project Formulation - Design of a Chemical Installation.- Appendices.- Workshop Program.- List of Contributors.
17 citations
14 citations
TL;DR: The combination of simulation and optimization for probabilistic models with continuous decision variables is discussed and the stochastic quasigradient method which is a well known technique in Stochastic optimization may also successfully applied for simulation-optimization problems.
Abstract: A major part of all simulation models contains a number of decision variables. For such models the problem of optimal decision arises in a natural way. The combination of simulation and optimization for probabilistic models with continuous decision variables is discussed in this paper. Several important techniques for solving the combined problem are presented. In particular the stochastic quasigradient method which is a well known technique in stochastic optimization may also successfully applied for simulation-optimization problems.
13 citations
11 citations
Proceedings Article•
01 Jan 1984
TL;DR: A water resource-constrained, land-use planning model suitable for use in the developing world, which allocates population among city subareas and specifies the type of sanitary service and level of water supply to be provided.
Abstract: Rapid urbanization in the Third World is exacerbating problems of water supply, flooding, and water pollution, as well as rendering more difficult the provision of such basic services as sanitation. Although fundamental to human welfare, these water-related issues seldom play a major role in guiding urban development. This paper is a description of a water resource-constrained, land-use planning model suitable for use in the developing world. The model allocates population among city subareas and specifies the type of sanitary service and level of water supply to be provided. The core of the model consists of a multiobjective optimization component linked to a rainfall-runoff simulation submodel. The program from the optimization model serves as input for the simulation of the flood regime under the given development conditions. Goal achievement is then evaluated, and reformulation of the optimization model is undertaken if desired. Specific applications include the determination of optimal residential-se...
TL;DR: In this article, a rational model for making engineering tradeoff decisions is presented, which is a hybrid from the fields of social welfare economics, communications, and operations research, and a solution methodology (vector optimization decision convergence algorithm or VODCA) is developed conceptually and mathematically.
Abstract: A rational model for making engineering tradeoff decisions is presented. The model is a hybrid from the fields of social welfare economics, communications, and operations research. A solution methodology (vector optimization decision convergence algorithm or VODCA) based on the economic model is developed conceptually and mathematically. The objective for developing VODCA was to improve the process for extracting relative value information about the objectives from the appropriate decision makers. This objective was accomplished by employing data filtering to increase the consistency of the relative value information and decrease the amount of information required. VODCA is applied to a simplified hypothetical tradeoff decision problem. Possible uses of multiple objective analysis concepts and the VODCA methodology in product-line development and market research are discussed.
01 Jan 1984
TL;DR: In this article, a trade-off analysis is carried out with respect to the active constraints, where the level of an active constraint is changed a little and then the optimization is done again to see how much the optimized objective will change.
Abstract: People who are not familiar with the multicriteria optimization theory often deal with multicriteria problems in the following way. One objective is taken as a single criteria to be optimized and some limits are set for the values of the other objectives. That is, the other objectives are treated as constraints in an ordinary optimization problem. After the optimization, a trade-off analysis may be carried out with respect to the active constraints. That is, the level of an active constraint — the limit of an objective — is changed a little and the optimization is done again to see how much the level of the optimized objective will change.
01 Oct 1984
TL;DR: In this article, the design of vibration isolation systems using multicriteria optimization techniques is considered using the integrated values of the square of the forces transmitted to the main mass and the relative displacement between the main masses and the base are taken as the performance indices.
Abstract: The design of vibration isolation systems is considered using multicriteria optimization techniques. The integrated values of the square of the force transmitted to the main mass and the square of the relative displacement between the main mass and the base are taken as the performance indices. The design of a three degrees-of-freedom isolation system with an exponentially decaying type of base disturbance is considered for illustration. Numerical results are obtained using the global criterion, utility function, bounded objective, lexicographic, goal programming, goal attainment and game theory methods. It is found that the game theory approach is superior in finding a better optimum solution with proper balance of the various objective functions.
TL;DR: This paper considers a finite serial multistage system where the measure of effectiveness of the system is a ratio of two return functions and shows that the optimal solution is a nondominated solution of the two criteria program.
TL;DR: In this article, sufficient conditions for the connectedness of certain efficient point sets of convex optimization problems with multiple or random objectives are provided, and known concepts of efficiency like functional efficiency and proper efficiency as well as efficiency with probability 1 (with respect to stochastic problems with fixed feasible domain) are treated.
Abstract: In this paper sufficient conditions for the connectedness of certain efficient point sets of convex optimization problems with multiple or random objectives are provided. Known concepts of efficiency like functional efficiency and proper efficiency (with respect to vector optimization problems) as well as efficiency with probability 1 (with respect to stochastic optimization problems with fixed feasible domain) are treated. By this the number of objectives that have to be minimized on the feasible domain (a subset of a Banach-space) can be infinite also.
TL;DR: One interactive algorithm is proposed which permits us to overcome the well known difficulties when choosing the weghting matrices in the cost functional and a procedure to design low parameter sensitivity optimal regulators by using a bicriteria formulation is presented.
TL;DR: Water systems seem to be ideal for application of hierarchical methods and some chosen hierarchical concepts are surveyed and their applications to these systems are presented.
TL;DR: This effort describes the modeling and optimization of an application which utilizes computer aided MOOT for candidate policy evaluation prior to a production process for an airborne tactical missile where the pre-production decision depends on such attributes as reliability, cost, technical performance, and survivability.
TL;DR: In this article, the problem of optimal selection of cutting data for multi-pass manufacturing with given constraints on the cutting times is investigated, and a secondary optimization method is proposed to enforce management-type requirements for the optimization of data when manufacturing whole parts.
01 May 1984
TL;DR: The authors consider the case in which the lower level of a hierarchical system of decision makers is composed of a number of controllable subsystems, which could lead to a two-level system consisting of a regulatory center and one lower subsystem.
Abstract: This is the second of two papers dealing with mathematical methods that can be used to analyze hierarchical systems. In this paper, the authors consider the case in which the lower level of a hierarchical system of decision makers is composed of a number of controllable subsystems. If these subsystems are not bounded by common constraints then the analysis is reduced to that of a two-level system consisting of a regulatory center and one lower subsystem. Two types of control are discussed in this case: control of resource use and control through price setting. If, however, there are shared resource-type constraints then it is assumed that the subsystems choose cooperatively from the set of Pareto-optimal alternatives. The problem for the regulatory center is then to maximize its goal function over this set. A number of ways of solving this problem are proposed, and a computational algorithm is given.
TL;DR: Relations between efficient solutions to discrete multicriteria decision problems and optimal solutions to corresponding parametric single-criterion problems are analyzed.
Abstract: Relations between efficient solutions to discrete multicriteria decision problems and optimal solutions to corresponding parametric single-criterion problems are analyzed. In particular we consider such decision problems where one criterion is a sum function while the remaining criteria are bottleneck functions. In this case the multi-criteria problem can be reduced to a linear parametric sum problem. Depending upon the nature of its feasible set, the sum problem may belong to a class of computationally tractable optimization problems, By solving it parametrically we can then generate the set of all efficient solutions to the original multicriteria problem.
01 Jan 1984
TL;DR: In this paper, duality results for multi-objective optimization problems were derived by usual separation techniques, which generalizes known results in view of the applications to m.o. problems, which are presented in Section 3.
Abstract: This paper is concerned with duality results for multi objective (m.o.) optimization problems. The core of the paper is a duality theorem derived by usual separation techniques. This theorem generalizes known results in view of the applications to m.o. problems, which are presented in Section 3.