Computer simulation using a degradation model is the most effective method to investigate the degradation behaviors of poly lactic acid (PLA). Various kinetic parameters are introduced into numerous degradation models to achieve the best simulation result. Nevertheless, massive possibilities of different parameter combinations limit the application of the enumeration algorithm, while the nonlinear relationship between the kinetic parameters and the degradation behaviors of PLA indicates that the ordinary parameter search algorithms cannot do well in the parameter optimization. A genetic algorithm (GA) with a small population size is proposed and utilized to optimize the kinetic parameters of the cellular automaton (CA) simulation in the present work. The optimal result indicates that the presented GA can realize the parameter optimization of the CA degradation model. The elitist tournament selection operation can speed up the optimization process. The algorithm can be executed as a single-stage algorithm alone or applied as a multistage algorithm according to various solution objects and corresponding fitness functions. Moreover, the algorithm can be hybridized with other traditional search methods such as binary search or local enumeration search to achieve a balance between accuracy and search speed.
Keywords: PLA; cellular automaton; degradation; elitist tournament selection; genetic algorithm.