Optimum Seeking Methods (several variables)
For simplicity, our discussion is confined to the case of two variables only. Most of the optimum seeking methods for two variables can be easily generalized to the case of more than two variables, but the number of required trials grows very fast when the number of variables is increased. Thus the use of optimum seeking methods for several variables should be avoided as much as possible, even though there are always many factors that influence an industrial production process. It is better to grasp the principal factors and to use an optimum seeking method on one or two variables so as to obtain a better production technology for these factors, and then use the same method to improve the production technologies with respect to the next one or two factors, and so on, until a satisfactory result is obtained. It seems that this is more reliable and reasonable than the direct use of an optimum seeking method for several variables. We refer back to section 3.1 for some of the real problems we have tackled.
KeywordsConvex Body Maximum Point Equilateral Triangle Unimodal Function Bisection Method
Unable to display preview. Download preview PDF.