In an ordinary optimum design process, designers are often required to express a design problem as a mathematical model with a single objective. However, a number of objectives are usually found in structural designs, and it may then be tedious and difficult for designers to choose just one objective among them. In such cases, it will be natural and reasonable to formulate a multi-objective optimization problem. Recently, many studies have been made on interactive multi-criteria optimum design problems. In conventional methods, designers have been forced to make important and difficult decisions concerning the "trade-off ratio", "marginal rates of substitution", "surrogate worth trade-off" and so on, which may give much information in the optimization. To reduce the load of the designer in both the decision-making process and formulation, we have done a series of studies on qualitative optimization which based on qualitative and fuzzy reasoning. In this paper, we review these studies, clarify the definition of the concept of qualitative optimization and then examine the efficiency of the algorithm. From numerical examples of the optimum designs of a parabolic antenna, it is shown that the proposed method is applicable and effective for optimizations of complex structural systems by qualitative understanding, and can well support decision-making by the designers. In the layout optimization problems of a parabolic antenna, we show an extension of the method to the primary stage of the design process, that is, the conceptual or fundamental design, which has seemed difficult to handle in the conventional optimization method.
ASJC Scopus subject areas
- Computer Science Applications
- Computational Mechanics