Multiple criteria model for allocating new medical robotic devices to treatment centres

Abstract

This paper presents a multi-criteria methodology for allocating a small number of novel robotic devices to a set of potential treatment centres in the South of the UK. The robotic devices provide a novel means of diagnosis and treatment of prostate cancer and are currently in the design phase of an EU-sponsored research project. The aim of the project is to provide enhanced access to the new prostate cancer diagnosis and treatment technology. Hence, location and allocation decisions need to be optimised to facilitate this goal. A fuzzy logic based methodology based on measures of prevalence and current access to treatment centres is developed in order to identify a set of priority locales (postcode districts) for improved access. An extended goal programming facility location model is then formulated which incorporates both priority and general postcode access measures. The p-median, p-centre and coverage variants of facility location are modelled via a set of six goals. A comprehensive weight sensitivity analysis is used to generate 25 distinct solutions across the six-dimensional objective space. Analysis of the results with respect to methodology, location and allocation decisions, number of robotic devices and trade-offs between goals is undertaken and conclusions are drawn.