Department of Engineering / News / Cambridge engineers use industrial modelling techniques to help Addenbrooke’s manage COVID-19 care

Department of Engineering

Cambridge engineers use industrial modelling techniques to help Addenbrooke’s manage COVID-19 care

Cambridge engineers use industrial modelling techniques to help Addenbrooke’s manage COVID-19 care

Addenbrooke's Hospital site

Modelling tools originally designed to improve the efficiency of factories are being used by Cambridge engineers to help Addenbrooke’s Hospital manage the COVID-19 emergency.

Instead of production lines, we’re now looking at hospital wards, and rather than products or raw materials, we are examining the flow of patients and supplies.

Professor Duncan McFarlane

The work enhances the hospital’s own modelling, and provides insight into how day-to-day activities might be affected by a  rise in patient numbers in the coming weeks or months.

‘Discrete event simulations’ have been worked up by a team from the University’s Department of Engineering to manage the flow of patients through hospital wards in the event of a surge in cases, and anticipate waiting times, bed availability, and equipment and staff shortages.

Lead on the simulation development, Dr Ajith Parlikad, of Cambridge’s Institute for Manufacturing (IfM), said: “It’s looking at the physical flow of patients and projecting admissions rates into the future – identifying where ‘bottlenecks’ might occur, and where the hospital might need to scale up beds, ventilators, oxygen and staff as part of their COVID-19-orientated activities.

“We started with a flow diagram of how we thought the hospital worked, then talked it through with the team at Addenbrooke’s. It was quite close to their own model, but we were able to factor in more details, such as ICU beds, ‘COVID-positive’ beds (patients with the virus who don’t require intensive care), and the initial checking and testing process when patients arrive – everything has a statistical distribution associated with it.”

As well as patient flow modelling, the Department is supporting Addenbrooke’s in a number of other ways. Industrial engineering students are volunteering their time to focus on the hospital’s oxygen supply, among other things how it might be replenished and filtered, and are also looking at how to model and optimise COVID-19 testing processes. Colleagues Tom Ridgman and Dr Florian Urmetzer are co-ordinating the volunteer student group.

In addition, further work is beginning on modelling that will help the hospital better understand staffing level availability during disruptions such as the COVID-19 outbreak.

Duncan McFarlane, Professor in Industrial Information Engineering, said: “Two weeks ago we knew very little about hospital operations, but with close input from the hospital, we’ve been able to pick from a series of industrial techniques and apply the most useful ones to this new setting. Instead of production lines, we’re now looking at hospital wards, and rather than products or raw materials, we are examining the flow of patients and supplies.

“The support from the hospital has been extraordinary – especially given the level of pressure it has been operating under." 

Dr Ewen Cameron, Director of Improvement and Transformation at Cambridge University Hospitals, added: “At this time of unprecedented change for the NHS, our teams are working around the clock to set up innovative ways of working to best care for patients and protect our staff.

“The hospital looks very different now to a few weeks ago, and we remain open to additional ideas on how to manage this crisis as best we can. New challenges require new ways of thinking, and we are hugely grateful to the Institute for Manufacturing (IfM) for offering their expertise to help us beat the virus.”

How you can support Cambridge's COVID-19 research effort

Donate to support COVID-19 research at Cambridge.

The text in this work is licensed under a Creative Commons Attribution 4.0 International License. If you use this content on your site please link back to this page. For image use please see separate credits above.