Brooks' law revisited: a system dynamics approach

Abstract

Brooks' law says that adding manpower to a late software project makes it later. F.P. Brooks Jr. (1995) developed the law through observation of many projects and derived the generalization. His explanation was quite reasonable and convincing. However, it becomes a debilitating statement to any software project manager who is faced with a late project. This paper presents an in-depth study of Brooks' law using the system dynamics simulation approach. Unlike the first-order approximation of Brooks' law, we found some interesting results that can benefit practicing project managers. We found that adding people to a late project will always increase its cost, but the project may not always be late. Only under a certain degree of sequential constraints among the project tasks will the project be delayed. To investigate the impact of sequential constraints, we simulated numerous task conditions and found that there is a time-line T for each project which means that, if enough manpower is added before T, the project still can finish before the scheduled deadline. If the manpower adjustment is after T, then the project will definitely be late. In this way, Brooks' law is revised and turned into a useful guiding principle to benefit software development.