"A doctor doesn't have to write a duty rota"

Wuppertal professor Sebastian Rachuba develops models to support complex planning decisions in the healthcare sector

"We want to provide decision support," says Sebastian Rachuba, whose research focuses on a wide range of problems in the field of healthcare operations management. But what does that actually mean? "The English term operations actually refers to activities related to the provision of services," explains the junior professor, who has been working at the Chair of Production and Logistics in the Faculty of Business Administration and Economics at the University of Wuppertal since 2017. "In the healthcare sector, which is the health care part, we look at the original service provision processes. In other words, what are the workflows of the doctors? How does the duty rota work? Who has to do which errands, etc." The task of management is then to coordinate the entire system's processes. Whether it's changes to regulations, doctor scheduling, holiday requests or the deployment of part-time staff, the list of tasks could go on and on and is so complex that it is difficult to manage without an appropriately automated basic framework. And that's where the scientist would like to help, because "it's still often done with pen and paper," he explains, which would be easier in the age of digitalisation. Rachuba emphasises that this task can, of course, only be carried out in cooperation between the institutions and the university. "You have to develop it together with the users."

Optimisation model and simulation model

Rachuba works with two different models; the optimisation model for decision support and the simulation model, which can map processes. "With optimisation models, I have a lot of options to choose from and still have to take a lot of restrictions into account," he explains, using the example of operating theatre occupancy. "If I have a list of patients, many operating theatres and many days, then the whole thing becomes very complex very quickly. Then there are so many possible combinations, but how does the decision-maker actually know which one is good? Should Mr Meier go to room 1 or room 2 on Monday and so on? But I can try to make a suggestion with an optimisation model by asking: Find a plan where I work as little overtime as possible, but still treat as many patients as possible. That might not suit every surgeon, but it might make sense for the system as a whole." His optimisation models therefore develop solutions that minimise costs, optimise time or maximise capacity utilisation, depending on the requirements.
Rachuba's simulation models usually start afterwards. In the case of the emergency department, existing processes can also be visualised. He has already used this successfully in collaboration with a hospital in England. A new treatment pathway due to a new healthcare regulation, which was implemented for potential heart attack patients with a repeat blood test, had to be changed during operation. "So there were new medical findings that led to a guideline being issued. The question was, what would happen if a hospital of this size were to carry out this two-blood test procedure now? Would the processes still work as they do at the moment? Together with the users, we analysed that the process would now have to look different."

The emergency room: an example

"We try to take what happens in real life, abstract it, formalise it so that we can carry out experiments." Using the example of a simulation model, "we can model how an emergency room works". We then try out a different shift system or simulate the changes brought about by new regulations during ongoing operations. Possible change processes are developed together with the users at a round table. In this protected space, possibilities can be worked out, adjustments can be made and options can be played through in a what-if analysis to improve the process. With regard to overloaded emergency departments, a simulation can show, among other things, the sensible deployment of nursing staff, which would not necessarily have to be increased, but would have to be deployed differently in terms of time. "That's always the exciting part and sometimes also the most time-consuming. Then you sit down with the doctors and nurses to understand what is happening in the emergency department."

Nursing staff absences due to illness can also be taken into account in monthly or annual rota models. "Then I get a duty rota that may be a little overstaffed when everyone is there." The users ultimately decide whether they want this," says Rachuba. "In any case, there would then also be time to talk to the patient a little longer or to use the time for documentation."

Efficient planning of operating theatre occupancy

As early as his doctorate, the native of Herne, Germany, looked at how different objectives can be taken into account simultaneously. For the efficient planning of operating theatre occupancy, he examines the needs of different target groups. The patient wants to spend as little time as possible in hospital, the hospital management wants the individual treatment to be carried out quickly and the staff are interested in smooth processes without overtime. "It's often the case," explains Rachuba, "that general surgery has theatre 1 on Mondays and orthopaedics has theatre 2, but a general surgery patient is rarely operated on in the orthopaedics operating theatre." These old patterns need to be scrutinised. "If you blur things like this more, you can utilise the times much more sensibly. I might have a patient in general surgery whose operation will probably only take an hour and the theatre in orthopaedics is still free for two hours. I could possibly also use the theatre for the general surgery operation, so I would have one more operating theatre, the staff would be kept busy and I still wouldn't have any overtime. It all fits."
The economics graduate is well aware that the theories can fail because of department heads. If the head physicians get along with each other, "that's an idea," he says, "we can then use the optimisation models to provide a kind of template for the week in the operating theatre to support decision-making." With a simulation model, we could even visualise the annual operating theatre volume of a hospital and then analyse whether the resources are being used sensibly. To put it bluntly: "I waste less operating theatre time and staff time, and people get to the table quickly."

Where should ambulance stations be?

The siren is blaring, the ambulance is on its way. People hear and experience this every day in city traffic. Sebastian Rachuba scrutinises the efficiency of this too. "The basic rule is always this: Ambulances should be positioned where the demand is." Sounds simple enough, but the reality is different. The focus of operations shifts over the course of the day, which also means that there are not always the same number of vehicles at the stations. "In a study with colleagues at the University of Bochum, we have already analysed the option of not stationing ambulances exclusively at rescue stations," reports Rachuba. "We analysed various locations, e.g. hospitals, schools or other public buildings, so that a toilet or social rooms would also be available. Essentially, these were locations where the ambulances would possibly be much better positioned because they would then be closer to the centres of operations," he sums up. Replenishing used equipment could also not be done exclusively at the rescue centres. "Flexible positioning certainly offers many more possibilities, which would have to be discussed with the users." He can also use a simulation to visualise all of this and show how much time would be saved if, for example, the vehicles did not have to keep returning to their location.
 

A hospital in Solingen wants change

Rachuba worked for some time as a consultant at a hospital in Solingen. Existing contacts from this time with leading doctors in the hospital are very valuable to him. He can discuss his ideas with the head of the emergency department and also gains an insight into the views of the professionals. The realisation that many things are not feasible for the medical and nursing staff gives rise to a desire for improvement. The question of whether we are really as good as the professional association prescribes leads to the first decisive step of analysing ourselves. This is where Rachuba can start with his models. The opportunities for change have been recognised in Solingen. Rachuba notes: "There has to be the will to dare to tackle something as unpleasant as change processes. In hospitals, specialised staff have not been trained to improve logistical processes or solve complex planning tasks. A doctor doesn't have to write a duty rota. That could be automated. What's more, these are not medical tasks. They can be done by people who are better at it, who have time for it and who are familiar with it. Here too, dialogue is of course essential for the success of this decision support. If processes run more smoothly and decisions are made easier because they come automatically from the computer, it becomes simpler."
The digital hospital of the future will make work easier, as decision-makers from hospitals recently discussed at the second Hospital Innovation Forum at Deutscher Ärzteverlag in Cologne. Artificial intelligence will make work processes easier, but also requires staff to be trained accordingly. Digitised triage in the emergency department, i.e. prioritising the medical treatment of patients, or web-based duty scheduling promise to increase employee job satisfaction.
Optimisation and simulation models, such as those developed by Sebastian Rachuba, could effectively support this development.

Uwe Blass (interview from 18.09.2019)
 

Sebastian Rachuba studied economics at the Ruhr University Bochum, where he also completed his doctorate in 2013. He then worked for a year as a consultant to the Medical Director at Solingen Municipal Hospital. From 2014 to 2017, the native of Herne was a postdoc at the Medical School of the University of Exeter. Since 2017, he has been the new Junior Professor of Business Administration, in particular Operations Management, at the Schumpeter School of Business and Economics at the University of Wuppertal.