Budget allocation optimization in a complex multi-project environment

Abstract

Most organizations are complex environments, where multiple projects interact with each other in many different ways. Projects can collaborate either serially or in parallel, yet at other times they compete with each other to test different approaches for the same goal. In such organizations, it is not uncommon for the wrong project to be funded or a right project to be under funded apparently for all the right reasons. In turn these project failures can cause big problems to their dependent projects, even having catastrophic impacts to the whole organization. Given a limited budget, it is critical to the organization's success to strategically optimize the set of projects that are initiated, and for these projects to be able to depend on their selected project dependencies or collaborators. Conceptually this is obvious, but in practice to identify the right set of projects and to allocate the right amount of budget is very challenging. We propose an approach to tackle the problem from a risk management perspective: first model each project as a chain of executable tasks and model the organization as a network of tasks; then optimize the budget allocation from the system's perspective.