Probabilistic models of project management with design implications

As project management has come of age, there has been an increasing awareness of the fact that the greater the inherent difficulty of the R&D activity, the more likely it becomes that the project will encounter cost and schedule overruns. Some authors have even gone so far as to call for `built-in´ allowances for overruns. It is the purpose of this paper to point out not only that such a scheme is necessary, but also to show the magnitude of the required allowances and their sensitivity to the distribution of risk associated with the project.