Stacking sequence optimization for OHC strength of laminated composite by tabu search using the Bayesian surrogate model in the lamination parameter space

This study proposes an efficient stacking sequence optimization method using a tabu search alogorithm together with a zigzag method to efficiently find the optimal stacking sequence of composite laminate that maximizes open-hole compression (OHC) strength while considering empirical stacking constraints. The proposed method improves the search performance for the optimal stacking sequence within the Bayesian optimization framework. This framework determines the optimal stacking sequence that maximizes strength based on the experimental results. The authors propose that strength be approximated by a Gaussian process in the stacking parameter space. The zigzag method, a previously proposed method, is named for the search path in the lamination parameter space that swaps the locations of two plies or inserts a ply to improve the strength. However, since the zigzag method is a local search method, it may fall into a local optimum depending on the initial design. This study reduces the dependency on the initial designs by introducing the tabu search.