Pengyi Shi

Associate Professor of Operations Management
Krannert School of Management
Purdue University
Email: shi178 AT purdue DOT edu

[CV]

Google Scholar Page
    Pengyi Shi


About me

I am an Associate Professor of Operations Management at the Krannert School of Management, Purdue University. I am an affiliate faculty of the Regenstrief Center for Healthcare Engineering and the Integrative Data Science Initiative.

I obtained my Ph.D. degree from the School of Industrial and Systems Engineering at Georgia Institute of Technology in December 2013. I received my B.Eng. in Applied Mathematics and Mechanics and B.A. in Economics (double major) from Peking University, China, in June 2007. My Ph.D. advisors are Jim Dai and Pinar Keskinocak. Here is a copy of my Ph.D. thesis [pdf].


Research

My research focuses on building data-driven, high fidelity models and developing predictive and prescriptive analytics to support decisions making under uncertainty in healthcare and service systems.

One of my main research stream is to develop patient flow models to improve hospital operations and patient outcomes. This stream of research has been implmeneted as tools for supporting inpatient discharge management and for supporting COVID-19 response in the hospital systems in Indiana. Recently, I have started working on developing predictive and operations tool for criminal justice system and its interface with substance use abuse.

My research methodologies include stochastic models, queueing theory, Markov decision process, machine learning, reinforcement learning, and online learning. See my publications and here for a guest lecture I gave on applying reinforcement learning to tackle large-scale decision problems in hospital operations and criminal justice system. I am currently serving as an Associate Editor of Operations Research, Health Care Management Science, Service Science, and guest AE for Naval Research Logistics.

I strive to develop high impact research and tackle real-world problems. I have been collaborating with practitioners and faculty members from different healthcare organizations, public sectors, and goverment agencies, publishing on medical journals. I was selected into the first cohort of Purdue Societal Impact fellows. As the PI, I received the 2021 Engaged Scholarship Research/Creative Activities Grants and the 2022 Shah Lab Global Innovation Lab Seed Grant. I have worked extensively with large-scale data sets from hospitals and public sector organizations in the US, Singapore, and China.

I have had the pleasure of working closely with PhD students and undergraduate students; see my research team. Graduate students with solid theoretical background could check out this research opportunity Deep-learning Enhanced Healthcare Modeling and Optimization. Purdue undergraduate students are encouraged to apply through my DURI project.


Publication

Selected Working Papers

denotes industry/medical collaborator; underline denotes student
    • Stochastic Models and Control

  1. J. Chen, J. Dong, P. Shi
    Optimal Routing under Demand Surges: The Value of Future Arrival Rates. [Abstract and full paper]
    • Finalist, 2021 INFORMS Service Science IBM Best Student Paper Competition

  2. Y. Pan, P. Shi
    Refined Mean-Field Approximation for Discrete-Time Queueing Networks with Blocking. [Draft]
    • Finalist, 2021 INFORMS Undergraduate OR Prize

  3. Y. Liu, P. Shi, J. E. Helm, M. P. Van Oyen, L. Ying, and T. Hucshka
    Coordinated Care: Capacity Allocation to Improve Itinerary Completion in Queueing Networks. [Abstract and full paper]

  4. J. Dong, P. Shi, F. Zheng, and X. Jin
    Structural Estimation of Load Balancing Behavior in Inpatient Ward Networks.

  5. J. Dong, P. Shi, F. Zheng, and X. Jin
    Off-service Placement in Inpatient Ward Network: Resource Pooling versus Service Slowdown. [Abstract and full paper]
    • Second place, College of Healthcare Operations Management Best Paper Award, POMS, 2020
    • Selected to 2019 MSOM Conference Healthcare SIG
    • Machine Learning and Reinforcement Learning

  6. X. Liu, B. Li, P. Shi, L. Ying
    An Efficient Pessimistic-Optimistic Algorithm for Constrained Linear Bandits. [Abstract and full paper]
    • Preliminary accepted at NeurIPS 2021 (26% acceptance rate)
    • Selected to RLNQ 2021 Workshop

  7. T. Jang, P. Shi, and X. Wang
    Group-Aware Threshold Adaptation for Fair Classification. [Abstract and full paper]
    • Preliminary accepted at Thirty-Sixth AAAI Conference (15% acceptance rate)

  8. X. Chen, P. Shi, and X. Wang
    Data Pooling for Reinforcement Learning with Applications to Healthcare Intervention.
    • Preliminary selected to ICML RL4RL Workshop, Spotlight Talk [Talk Video]

  9. H. Bastani, P. Shi
    Proceed with Care: Integrating Predictive Analytics with Patient Decision-Making.
    Invited book chapter for Modeling for Health: Making Changes. [Draft]

Papers Accepted and Published in Operations Journals

denotes industry/medical collaborator; underline denotes student
  1. P. Shi, J. E. Helm, C. Chen, J. Lim, R. Parker, T. Troy, and J. Cecil
    Operations (Management) Warp Speed: Parsimonious Design for Rapid Deployment of Hospital Prediction and Decision Support Framework during a Pandemic.
    Forthcoming, POM special issue on Managing Pandemics: A POM Perspective. [Abstract and full paper]
    • I am the sole academic researcher in developing the adaptive workload/census prediction, which is a core component of the integrated tool for COVID-19 surge planning.
    • This tool has been implemented in IU Health (the largest hospital systems in Indiana) since April 2020 to support their COVID-19 response. Below are several selected news release on our work:

  2. J. Chen, J. Dong, P. Shi
    A Survey on Skill-Based Routing with Applications to Service Operations Management.
    Queueing Systems 2020; 4:1-30.

  3. P. Shi, J. E. Helm, S.H. Heese, and A. Mitchell
    An Operational Framework for the Adoption and Integration of New Diagnostic Tests.
    Production and Operations Management. Forthcoming. [Abstract and full paper]

  4. P. Shi, J. E. Helm, J. Deglise-Hawkinson, and J. Pan
    Timing it Right: Balancing Inpatient Congestion versus Readmission Risk at Discharge.
    Operations Research. Forthcoming. [Abstract and full paper]
    • First place, MSOM Responsible Research in OM Award, INFORMS MSOM, 2021
    • First place, Pierskalla Best Paper Award, INFORMS HAS, 2018
    • Second place, Innovative Applications in Analytics Award, INFORMS Analytics, 2020
    • Second place, College of Healthcare Operations Management Best Paper Award, POMS, 2019
    • Selected to 2019 CHOM Mini Conference Showcase Presentations
    • Selected to 2019 MSOM Conference Healthcare SIG
    • Pilot implementation of the tool at a local hospital in Indiana. Also see a video on this research and a newsletter article.

  5. J. G. Dai, P. Shi
    Recent Modeling and Analytical Advances in Hospital Inpatient Flow Management.
    Production and Operations Management. Forthcoming. [Abstract and full paper]
    • Invited paper for special issue on Frontier Analytic Modeling and Methods for OM.

  6. A. Alaeddini, J. E. Helm, P. Shi, and S. Faruqui
    An integrated framework for reducing hospital readmissions using risk trajectories characterization.
    IISE Transactions on Healthcare Systems Engineering. 2019; 9(2):172-185.
    • Former title: "A Prediction Model for Patient Readmission Risks with Kernel Principle Component Analysis."

  7. J. G. Dai, P. Shi
    Inpatient Bed Overflow: An Approximate Dynamic Programming Approach.
    Manufacturing and Service Operations Management. 2019; 21(4):894-911. [Abstract and full paper]

  8. J. Feng, P. Shi
    Steady-state Diffusion Approximations for Discrete-time Queue in Hospital Inpatient Flow Management.
    Naval Research Logistics. 2018; 65(1):26-65. [Abstract and full paper]

  9. J. G. Dai, P. Shi
    A Two-Time-Scale Approach to Time-Varying Queues for Hospital Inpatient Flow Management.
    Operations Research. 2017; 65(2):514-36. [Abstract and full paper]

  10. P. Shi, M. Chou, J. G. Dai, D. Ding, and J. Sim
    Models and Insights for Hospital Inpatient Operations: Time-Dependent ED Boarding Time.
    Management Science. 2016; 62(1):1-28. [Main Paper] [Online Supplement]
    • Former title: "Hospital Inpatient Operations: Mathematical Models and Managerial Insights."

Other Conference Publication

denotes industry/medical collaborator; underline denotes student
  1. Z. Zhang, P. Shi, A. Ward
    Routing for Fairness and Efficiency in a Queueing Model with Reentry and Continuous Customer Classes.
    Proceedings of the 2022 American Control Conference (ACC). [Full paper]

  2. I. Attari, P. Crain, P. Shi, J. Helm, N. Adams
    A Simulation Analysis Of Analytics-driven Community-based Re-integration Programs.
    Proceedings of the 2021 Winter Simulation Conference.

  3. Y. Pan, Z. Xu, et al. P. Shi, H. Pan, K. Yang, S. Wu
    A High-fidelity, Machine-learning Enhanced Queueing Network Simulation Model for Hospital Ultrasound Operations.
    Proceedings of the 2021 Winter Simulation Conference. [Abstract and full paper]

Medical Publication

denotes industry/medical collaborator; underline denotes student
  1. F. Dexter, R. H. Epstein, P. Shi
    Proportions of Outpatients Discharged the Same or the Following Day are Sufficient Data to Guide the Assessment of the Cases.
    Cureus. 2021; 13 (3).

  2. F. Dexter, R. H. Epstein, P. Shi
    Forecasting the Probability That Patient Will Remain in the Hospital Overnight Versus Have a Length of Stay of 2 or More Days.
    Cureus. 2020; 12(10): e10847.

  3. P. Shi, J. Yan, P. Keskinocak, A. L Shane, J. L Swann
    The Impact of Opening Dedicated Clinics on Disease Transmission during an Influenza Pandemic.
    PLoS ONE. 2020; 15(8): e0236455.

  4. P. Shi, F. Dexter, R. H. Epstein
    Comparing Policies for Case Scheduling Within One Day of Surgery by Markov Chain Models.
    Anesthesia & Analgesia. 2016; 122(2):526-38.

  5. F. Dexter, P. Shi, R. H. Epstein
    Descriptive study of case scheduling and cancellations within one week of the day of surgery.
    Anesthesia & Analgesia. 2012; 115(5): 1188-95.

  6. P. Shi, P. Keskinocak, J. L. Swann, B. Y. Lee
    The Impact of Mass Gatherings and Holiday Travelling on the Course of an Influenza Pandemic: A Computational Model.
    BMC Public Health. 2010; 10: 778.

  7. P. Shi, P. Keskinocak, J. L. Swann, B. Y. Lee
    Modeling Seasonality and Viral Mutation to Predict the Course of an Influenza Pandemic.
    Epidemiology and Infection. 2010; pp. 1-10.

Other Papers

  • P. Shi, J. G. Dai, D. Ding, J. Ang, M. Chou, X. Jin and J. Sim
    Patient Flow from Emergency Department to Inpatient Wards: Empirical Observations from a Singaporean Hospital.
    This is a 69-page document on a comprehensive empirical study of inpatient flow management. [pdf]

Contact Information

403 W State St Kran 472
Krannert School of Management
Purdue University
West Lafayette, IN 47907
Office phone: (765) 494-0458
Email: shi178 at purdue.edu