Multivariate analysis: techniques for educational and psychological research, 2nd ed.
Multivariate analysis: techniques for educational and psychological research, 2nd ed.
ACM Computing Surveys (CSUR)
Quality Engineering Using Robust Design
Quality Engineering Using Robust Design
Optimizing Engineering Designs
Optimizing Engineering Designs
Muiltiobjective optimization using nondominated sorting in genetic algorithms
Evolutionary Computation
Genetic algorithms with a robust solution searching scheme
IEEE Transactions on Evolutionary Computation
Worst-case tolerance design and quality assurance via genetic algorithms
Journal of Optimization Theory and Applications
Formal engineering design synthesis
Open-ended robust design of analog filters using genetic programming
GECCO '05 Proceedings of the 7th annual conference on Genetic and evolutionary computation
Optimizing back-propagation networks via a calibrated heuristic algorithm with an orthogonal array
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
The goal of robust design is to develop stableproducts that exhibit minimum sensitivity to uncontrollablevariations. The main drawback of many quality engineeringapproaches, including Taguchi's ideology, is that they cannotefficiently handle presence of several often conflicting objectivesand constraints that occur in various design environments.Classical vector optimization and multiobjective genetic algorithmsoffer numerous techniques for simultaneous optimization of multipleresponses, but they have not addressed the central quality controlactivities of tolerance design and parameter optimization. Due totheir ability to search populations of candidate designs in parallelwithout assumptions of continuity, unimodality or convexity ofunderlying objectives, genetic algorithms are an especially viabletool for off-line quality control.In this paper we introduce a new methodology which integrates keyconcepts from diverse fields of robust design, multiobjectiveoptimization and genetic algorithms. The genetic algorithm developedin this work applies natural genetic operators of reproduction,crossover and mutation to evolve populations of hyper-rectangulardesign regions while simultaneously reducing the sensitivity of thegenerated designs to uncontrollable variations. The improvement inquality of successive generations of designs is achieved byconducting orthogonal array experiments as to increase the averagesignal-to-noise ratio of a pool of candidate designs from onegeneration to the next.