Using a Sliding-Frame Approach for Scheduling Large and Complex IT Projects

Abstract

Many IT projects and software development projects are very complex and sophisticated involving a large coordinated team. Such projects are a constant part of the operations of software companies such as Microsoft, SAP, Oracle, Google, Yahoo, IBM, and others. Many other companies carry large software projects as part of their IT operations. As a result of the size and complexity of such projects, a rolling horizon approach for their planning and management is not only plausible but also desirable. For large projects, traditional project scheduling techniques cannot provide an optimal and timely solution to minimum project duration under precedence and resource constraints. This paper proposes a technique that allows utilizing non-polynomial (NP) algorithms in a heuristic manner – generating high quality schedules in reasonable time. This approach can be applied efficiently for solving most project scheduling problems. The main advantage of this approach is its ability to dissect the original problem into small sub-problems of a controllable size to which exact techniques can be applied. Thus, it neutralizes the complexity of the applied algorithms (and their non-polynomial growth). Moreover, the same technique could be used for implementing a rolling-horizon approach in project planning and management.