Computer experiments have attracted a broad attention in many fields of science and technology.The premises in most literature assume that all the input variables are quantitative.However,researchers often encounter computer experiments involving mixed input variables: both quantitative and qualitative variables.In this paper,we first propose a new type of design,called clustered-sliced Latin hypercube design,which is a special sliced Latin hypercube design with points clustered in the design region.Using such designs for computer experiments with qualitative and quantitative variables helps to capture the correlations between responses corresponding to different level-combinations of the qualitative variables.Further,associated with newly proposed designs,we develop an adaptive analysis strategy for computer experiments with qualitative and quantitative variables,which automatically selects useful information from all auxiliary responses to increase the prediction accuracy of the target response.This strategy is shown to be effective through several numerical examples including a real example from the food engineering literature.