Statistical experimental designs were used to optimize the process of phenol degradation by Candida tropicalis Z-04, isolated from phenol-degrading aerobic granules. The most important factors influencing phenol degradation (p < 0.05), as identified by a two-level Plackett-Burman design with 11 variables, were yeast extract, phenol, inoculum size, and temperature. Steepest ascent method was undertaken to determine the optimal regions of these four significant factors. Central composite design (CCD) and response surface analysis were adopted to further investigate the mutual interactions between these variables and to identify their optimal values that would generate maximum phenol degradation. The analysis results indicated that interactions between yeast extract and temperature, phenol and temperature, inoculum size and temperature affected the response variable (phenol degradation) significantly. The predicted results showed that the maximum removal efficiency of phenol (99.10%) could be obtained under the optimum conditions of yeast extract 0.41 g/L, phenol 1.03 g/L, inoculum size 1.43% (V/V) and temperature 30.04℃. These predicted values were further verified by validation experiments. The excellent correlation between predicted and experimental values confirmed the validity and practicability of this statistical optimum strategy. This study indicated the excellent ability of C. tropicalis Z-04 in degrading high-strength phenol Optimal conditions obtained in this experiment laid a solid foundation for further use of this microorganism in the treatment of high-strength phenol effluents.