Q&A with new faculty member Xiang Zhong

Xiang Zhong

Q&A with new faculty member Xiang Zhong
Assistant Professor in ISE
Xiang Zhong came to UF from University of Wisconsin-Madison in fall 2016; below is condensed version of a discussion about her professional background and research interests–

What is your area of research?

My research interests have revolved around stochastic modeling and data analytics with implications in healthcare delivery systems. Essentially I’ve drawn upon probability theory, queing theory, and Markov models to model the CAD delivery processes in outpatient clinics or in hospital departments in order to characterize the patient flow and streamline the patient flow and also configure the staffing scheduling and do capacity planning.

We can make a comparison with production systems– so with production systems you’re processing a part through the whole production process. In healthcare systems, we view patients as a part, and essentially our objective is to process the patients through different procedures and receiving different types of care. We try to model the process, like how the patient is transferred from different care delivery processes. For example, if we look at a microsystem, like just a clinical visit, we model the flow– arrive to the clinic, served by a medical assistant, then treated by the physician, then leave. We aim to quantify such procedures, and identify the potential system bottlenecks, like what is the most important in such a process that will cause delay of the process, or incur any adverse outcome for the patient. We try to make improvements, for example improving the flow by designing different patient flows, and also the providers’ work flows in order to improve the process

What are the differences trying to streamline healthcare as opposed to manufacturing?

“The healthcare sector is more resistant to change. In the manufacturing domain, you are the consultant, and they want to listen to you advice. But in healthcare, the physician has absolute power to make any decisions. Also, the objective is different, because in manufacturing money is the objective. But in healthcare, safety comes first. So if you want to improve efficiency, you probably add the risk of incurring safety issues. You might suggest shortening a surgery time by 10 percent, then you can see 10 more patients a day, but they cannot guarantee that, because you don’t know what would happen if you skip some process.”

I started in manufacturing, with production systems. But we found similarities shared between the two systems. Healthcare has become a very important issue, and has greater societal impact– for example, people facing lack of physicians, and physicians who are over-burdened by those insurmountable tasks and healthcare expenses are skyrocketing. We were thinking it’s a very important area, and we believe some methodologies we have developed in the production system can be transferred to healthcare.

So how do you make sense of such an unpredictable system?

One of the challenges we are facing in modeling the healthcare system is it’s unlike manufacturing systems– they’re more fixed or deterministic. Here in the real health systems, you’re dealing with people, and there is such a substantial variability within the system. In terms of the methodological part, we need to capture those two variabilities.

This is just an example of how the mathematical system could work– if you were just observing the system, then collect the data, and analyze it, we probably don’t know the root cause or reason behind such a phenomenon, but using our analytical models, we can derive the solid roots of principles. Like what we were finding in the primary care clinic is we know that our providers work in teams- like a physician will work with a medical assistant. But in other clinics they might work with two nurses. Why is it like this? Which configuration is better? If you interview them, they don’t know why they do that. We found that the healthcare system is very inefficient. There is lots of room for improvement. There are multiple reasons, but the biggest issue is we’re human being; we make mistakes.

Why did you come to UF?

It’s a great opportunity to start my career. I think UF has a lot of great resources I can reach out to– for example, the interdisciplinary research environment. People in the whole College of Engineering collaborate extensively.