Choice-point nets: A discrete-event modelling technique for analyzing health care protocols

Every health care system employs a set of protocols to manage and reduce the impact of infectious disease whenever it appears within the population. Although a great deal of research has been conducted to determine when an outbreak is occurring, research pertaining to whether the response policies are effective is not easily located. Much of the difficulty lies in selecting an appropriate modelling mechanism. To correctly capture a protocol´s characteristics, a model must incorporate time and probability, manage large numbers of people and offer analysis that can answer the questions health care administrators will want to ask. Choice-point nets (CNs) are an augmented form of Petri net and offer just such an approach. The enabled transitions in CNs must fire according to their defined timing constraints based on a global clock. Once fired, the outcome of the transition is selected from a set of choices, each of which has a probability attached. Analysis can be performed by unravelling the net into an augmented reachability graph. It is shown how CNs can be employed to analyze outbreak management protocols within a long-term care facility.