The purpose of a phase I clinical trial is to determine the maximum tolerated dose(MTD) of a new drug. This paper investigates the performance of standard method, continual reassessment method and accelerated titration designs in phase I clinical trials. Especially we study the precision and safety at the MTD of these methods. We utilize hyperbolic tangent function and power function to define dose-toxicity model. For each method, expected toxicity rate at MTD is computed and compared with target toxicity probability. We also suggest some modifications of these methods and show some improvements in performance.