Improved chromosome encoding for genetic-algorithm-based assembly sequence planning

In order to ensure every feasible assembly sequence is included in the searching space of genetic algorithm, this paper proposes an improved genetic encoding method. In this method, every chromosome is encoded by a permutation of integer. Although the final form of chromosome is the same as the existing form, the decoding process is different. In our decoding process, every permutation of integer is transformed into an ordered set of all edges in the stability-directed graph, and then the ordered set of edges is transform into an ordered set of assembly states. The correctness of this method is proved, the combination of this method and the existing genetic algorithm is discussed, and an example is presented to show the effectiveness of our encoding method.