Pengyi Shi
Purdue University, IN·Engineering
Hospitals frequently run short of nurses and other clinical staff. The usual fixes, expensive temporary workers and extra shifts, raise costs and wear people out, and patient care suffers.
- Fields
Engineering
Mathematics & Statistics
Computer Science
- Topics
operations research
stochastic optimization
machine learning
- Grants
- 1 known grant, through Mar 2027
What they work on
- Build a decision-support system that pairs machine learning demand forecasts with stochastic optimization to recommend staffing assignments across multiple hospital facilities in real time.
- Make staffing recommendations adapt continuously to shifting patient demand, hospital conditions and workforce availability, cutting understaffing and reliance on temporary workers.
- Evaluate the technology's translation potential through I-Corps experiential learning and direct investigation of the healthcare industry ecosystem, including adjacent sectors such as emergency response and logistics.
More about the research
Because demand swings quickly and differs from one building to the next, deciding who works where and when is genuinely hard. This project asks whether forecasting plus mathematical optimization can make those decisions in real time across a whole hospital system.
Why it matters
Hospital managers facing chronic shortages could receive concrete, continually updated staffing recommendations, lowering understaffing and dependence on costly temporary labor. That may ease clinician burnout and support more consistent patient care. Organizations outside healthcare with volatile demand, such as emergency services and logistics operators, could use the same allocation methods.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Seyed Iravani
Northwestern University, IL·Engineering
Develop behavioral and analytical models of how customers react to delay announcements, combining stochastic dynamic programming and queueing theory with prospect theory and econometrics to decide what wait information to give and when.
Emily Tucker
Clemson University, SC·Engineering
Quantify how staffing strain, patient congestion, pandemic status and demographics relate to well-being and burnout in emergency department clinicians, using generalized linear mixed effects models and factor analysis on survey data from seven sites.
Michael Fu
University of Maryland, MD·Engineering
Develop an optimization strategy to identify the optimal personalized desensitization protocol, the pre-surgery removal of antibodies that would otherwise cause rejection of a donated kidney.
The school
Purdue University: 43 labs on PI Pages, $932M a year on research, 833 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.