Choosing a correct software process and controlling the process effectively has been the key to guarantee the success of a software project. This paper presents a novel approach for assessing and optimizing a software project process based on software risk assessment and control, which evaluates and optimizes software project process from the view of controlling the software risk. A model for optimizing software risk control is given, and then a discrete optimization algorithm based on dynamic programming is proposed. An example of using above method to solve a problem is also included in this paper. By changing afterwards control passively into an effective pre-action actively, this new method can greatly promote the possibility of success of software projects.