Displacement prediction of tunnel surrounding rock plays a significant role for safety estimation during tunnel construction. This paper presents an approach to use support vector machines (SVM) to predict tunnel surrounding rock displacement. A stepwise search is also introduced to optimize the parameters in SVM. The data of Fangtianchong tunnel is use to evaluate the proposed model. The comparison between artificial neural network (ANN) and SVM shows that SVM has a high-accuracy prediction than ANN. Results also show SVM seems to be a powerful tool for tunnel surrounding rock displacement prediction.

Job shop scheduling problem (JSP) plays a significant role for production management and combinatorial optimization. An improved artificial bee colony (IABC) algorithm with mutation operation is presented to solve JSP in this paper. The results for some benchmark problems reveal that IABC is effective and efficient compared to those of other approaches. IABC seems to be a powerful tool for optimizing job shop scheduling problem.