Hybrid Multi-Criteria Decision Support Tool for Proper and Safe Medical Device Maintenance and Disinvestment

Abstract

Proper management of medical device is mandatory for sustaining high quality patient care, proper clinical and financial governance, and reducing the risk of adverse effect. However, due to several reasons such as lack of operator training, maintenance capacity and inadequate knowledge regarding complex and sophisticated equipment, the employment of proper maintenance and disinvestment are at substandard level, especially in Ethiopia. Consequently, serious health risks and huge economic deficit are being observed. To solve this, different multicriteria decision analysis based maintenance management approach were proposed in the literature. However, to our knowledge the existing literature did not incorporate inclusive and optimum selection, and prioritizing criteria. In addition, most of the studies didn’t consider uncertainty and vague information gathered from the expert’s. Failure mode prioritization is mostly implemented in industrial and manufacturing field; but it is not yet been implemented for medical device. Thus, compressive and well organized model that incorporates maintenance strategy selection, medical device prioritization, failure prioritization and medical device disinvestment is required for proper medical device management. In this thesis, a hybrid multicriteria decision analysis model is proposed for selecting suitable maintenance strategies and medical device and failure mode prioritization. In order to execute the model, suitable criteria and attributes required for maintenance strategy, device and failure prioritization were identified from literatures and situational analysis. The identified criteria were examined by selected experts, meanwhile; the consistency of each expert was checked. Then, the weights of the criteria were calculated using Fuzzy analytical hierarchy process (AHP ).To prioritize medical devices, criteria -based device assessment result and computed criteria weights were used. The alternatives such as maintenance strategy and failure modes were ranked by implementing Fuzzy technique for order preference by similarity to ideal solution (TOPSIS). The rank of calculated weight and alternatives were compared and found to be in line with the literature findings. In addition, medical device disinvestment decision support system was also developed. Finally, web-based application was developed for ease of use for the developed models and to facilitate the models applicability to a clinical setting.