Minimally Invasive Surgery and Interventional Techniques
Department of Biomechanical Engineering - Delft University of Technology
Prof.dr. Jenny Dankelman
Predicting medical device failure in LMIC hospitals
27 November 2020

Predicting medical device failure, its impact on hospital’s mission, and optimal maintenance schedules for LMIC hospitals


Hospitals in many developing countries have medical equipment that is old. Most equipment is far beyond the life-expectancy of the device, and even beyond the end-of-life and end-of-service-life of the model. With increasing age the risk of device failure increases, and the difficulty and cost of repair increases as well, especially once a device is beyond end-of-service-life (spare parts no longer produced by the manufacturer). For those hospitals it would be valuable to have a model that predicts the risk of failure for their equipment so they know for each device how likely it is that it will fail. If this is combined by the probability of timely repair because of the availability of spare parts and other factors it could be assessed if this devices is likely to fail without option for timely repair.


If there would also be a means to factor in the importance of the device for the overall mission of the hospital the impact of this unrepairable failure on the mission of the hospital could be computed. It matters for instance if this device is the only device of this kind that the hospital owns, and if this device is for instance a diagnostic device that would prevent setting diagnosis and thus treatment of patients. It would mean a huge risk to the mission of the hospital and to the income generated. This information would be very important to decide when to replace certain equipment as well as on which interval maintenance has to be performed. Such models might exist for (medical) devices in western hospitals or industry, but those devices generally will be within the life-expectancy of the device, and within the end-of-life and end-of-service-life of the model.


The aim of this project is to define a model that can predict the probability of failure of a device and its expected impact on the income and mission of the hospital, and connect that to advise on optimal maintenance schedules. This model would also be a tool for the hospital to decide on planned replacement of equipment. It could be applied to the inventory of the hospital to flag risks in their inventory, and integrated in a computer program used to manage the inventory and maintenance.



Jenny Dankelman,

Arjan Knulst,


Photo album