With the increasement of software size, software process control is becoming more and more important. A software process is a framework for the tasks that are required to build high-quality software. A decades-long goal has been to find repeatable, predictable processes that improve productivity and quality.This thesis proposed a task-oriented software process control model for reducing software costs and risks, improving the efficiency of the software process without sacrificing software quality. Task-oriented project plan is made to partition the software process into a series of tasks. Then constraints of tasks are built with finite domain logic programming and scheduling solution is solved.Firstly, we discuss the background of software process modeling and related works. Then we propose TSPM which is short for a task-oriented software process control model. Organization's set of standard process, project definition process and project execution process are defined on which we get the project plan algorithm based on the definition. Finally, we discuss the control and improvement of the model.In addition, how to schedule the tasks in the TSPM is an NP problem, which is dynamic stochastic and complicated. After comparative analysis of various algorithms, constraint programming is chosen to model the problem. Experimental results show that CP works well. |