Evolutionary algorithms for design and control of material handling and manufacturing systems

The crucial goal of enhancing industrial productivity has led researchers to look for robust and efficient solutions to problems in production systems. Evolving technologies has also, led to an immediate demand for algorithms which can exploit these developments. During the last three decades there has been a growing interest in algorithms which rely on analogies to natural processes. The best known algorithms in this class include evolutionary programming, genetic algorithms, evolution strategies and neural networks. The emergence of massively parallel systems has made these inherently parallel algorithms of high practical interest. The advantages offered by these algorithms over other classical techniques has resulted in their wide acceptance. These algorithms have been applied for solving a large class of interesting problems, for which no efficient or reasonably fast algorithm exists. This thesis extends their usage to the domain of production research. Problems of high practical interest in the domain of production research are solved using a subclass of these algorithms i.e. those based on the principle of evolution. The problems include: the flowpath design of AGV systems and vehicle routing in a transportation system. Furthermore, a Genetic Based Machine Learning (GBML) system has been developed for optimal scheduling and control of a job shop.