The Project Portfolio Management Problem

We consider the Project Portfolio Management Problem (PPMP) in which a limited resource must be allocated among a set of candidate projects over time so as to maximize expected net present value. We formulate this problem as a dynamic program but conclude that this approach is too computationally complex to be of value in supporting real-world project management. So, we investigate the structural properties of the optimal solution to the PPMP and demonstrate that the solution reduces to a simple form under certain environmental conditions. This simplified policy, which we term the index policy, sequences projects according to a simple ratio and then allocates resource up to each project’s practical limit in the order given by this sequence. Through numerical tests we demonstrate that this policy performs robustly well on the general PPMP. Hence, we conclude that the index policy is a practical way to incorporate economic and timing issues into a multi-dimensional scoring model for addressing real-world project portfolio management situations.