Adaptive Learning for Successful Design
- 468 Downloads
In this chapter we want to evaluate which of many parameters (each is set at various possible levels) composing a design solution have the greatest likelihood of satisfying a given set of functional requirements. The design’s functional requirements are represented by a set of prespecified limits that determine where the output responses should fall. Adopting the probabilistic paradigm presented in Chapter 7 and the methodology provided in Chapter 8 for quantifying how well a proposed design satisfies the governing requirements (in probabilistic terms), we present a method for adaptive learning of successful designs that is based on the use of statistical experimental design and a stochastic search algorithm. In Chapter 19, we present a real industrial problem of designing a flexible manufacturing system that is solved based on the proposed algorithm.
Unable to display preview. Download preview PDF.
- 1.Ross, P. J., Taguchi Techniques for Quality Engineering, McGraw-Hill, 1988.Google Scholar