Generalised activity-on-the-node networks for managing uncertainty in projects

Abstract Activity networks, for example PERT , have long been used to plan and manage all kinds of projects within both academia and industry. They are, however, limited by their inflexible structure, and so they cannot explicitly identify and control potential risk points and uncertainties within projects. Generalised activity networks provide a more realistic means of controlling projects by identifying, and analysing, scope possibilities within project plans. Project managers are currently more familiar with the activity-on-the-node network planning techniques on which almost all project management software tools are based. In contrast, generalised activity networks have always been restricted to an activity-on-the-arrow representation. Activity-on-the-node representations have several advantages over activity-on-the-arrow representations, including the ability for logical dependency constraints to be applied directly. The paper introduces the first comprehensive generalised activity-on-the-node representation, thus providing a technique that can identify variability in the scope, cost, and duration of a project.