A whale optimization algorithm–based cellular automata model for urban expansion simulation

Cellular automata (CA) has proved to be effective and efficient in conducting urban expansion simulation. The generation of cell transition rules is a crucial step for a CA model. In this research, a whale optimization algorithm–based CA (WOA-CA) model was innovatively proposed. In the proposed model, a WOA was adapted to help mining the transition rules of the CA model, which was also evaluated and utilized in the case study of Guangzhou, simulating urban expansion from the year of 2000 to 2010. The experiment results demonstrated that the proposed model is effective and the simulation result is able to reach an overall accuracy of 92.16% with a Kappa coefficient of 0.744, and the value of Moran's I is also quite close to that of the actual urban expansion. In addition, the proposed model has also been compared with a few representative CA models, including multi-criteria evaluation-based CA (MCE-CA), artificial neural network-based CA (ANN-CA), bat algorithm-based CA (BA-CA), convolution neural network for united mining-based CA (UMCNN-CA), and gray wolf optimizer-based CA (GWO-CA). The comparison results showd that our proposed model outperforms all these models in terms of overall accuracy and computational efficiency.