Approximating finite resources: An approach based on MVA

Blocking is the phenomenon where a service request to a queue is momentarily stopped, but not lost, until the service becomes available again. Blocking is a difficult phenomenon to model analytically, because it creates strong interdependencies between system components. Mean Value Analysis (MVA) is one of the most appealing evaluation methodologies due to its low computational cost and ease of use. In this paper, an approximate MVA for Blocking After Service is presented that greatly outperforms previous results. The new algorithm is obtained by analysing interdependencies due to the blocking mechanism and by accordingly modifying the MVA equations. The proposed algorithm is tested over about sixty cases and error and computational cost analysis is presented. Finally, the method is applied to capacity planning for healthcare systems, where its simplicity and ease of use is demonstrated, making the case that it can be easily employed by non-experts.