Managing the development of large-scale software systems is a challenge to all software project managers due to the ever-increasing complexity inherent in the software development life cycle. In this paper, a formalism intended to capture the concurrent and iterative nature of software development, called Software Project Management Net (SPMNet) is proposed to model software development projects. By augmenting and applying genetic algorithms in an innovative manner by sharply reducing the solution-search space complexity, our SPMNet provides optimal or near-optimal solutions to the resource allocation and project scheduling problems. Results collected from simulation runs clearly support our approach for practical applications.