Predicting workload requirements for scheduled health care services, with an application to radiology departments.

A model is developed for predicting workload requirements for scheduled health care services. The model is then applied to an actual planning problem for a radiology department. The probability distribution of future workload is represented by the convolution of two families of random variables such that a compound Poisson process adequately describes workload requirements. The model developed herein can be applied to a wide assortment of capacity-expansion problems that are characterized by discrete demands (e.g. number of jobs) occurring in a given period of time, where the amount of time needed to complete each job is a continuous random variable.