A new dynamic CPM Optimal Algorithm base on constraints of scheduling network plan is proposed. The Structure of proposed algorithm is similar as the model integrated CPM with Monte Carlo simulation, so that the constraints including actives sequence constraints and date limitation constraints are considered together with CPM. The statistic result is provided by algorithm after calculating all of the trials. The critical path and secondary critical path critical path and their operable probability can be achieved under constraints from the result. Matlab 2006, Visual Studio 2005 and SQL Server 2000 are utilized to implement proposed algorithm. The realistic case study has verified the validity of proposed algorithm, which obtains CPM，duration，information of related conflicts and their statistic characteristic; Such information is helpful to improve the decision support.