Diwakar Gupta is a professor of Industrial & Systems Engineering at the University of Minnesota. He also holds a courtesy appointment as an affiliate senior member in the Health Services Research, Policy, and Administration Division of the School of Public Health. Gupta earned a PhD in Management Sciences from the University of Waterloo. His research focuses on healthcare delivery systems, state transportation agencies' operations, and supply chain and revenue management. Professor Gupta's research has been funded by a variety of federal and state agencies (e.g., DHHS, NSF, AHRQ, VHA, Mn/DOT, NSERC, SSHRC, and CHSRF), as well as companies, and his papers have appeared in all major journals in the field of operations research/management science. Diwakar has held a variety of editorial appointments, including co-editor-in-chief of the Flexible Services and Manufacturing journal.
Diwakar’s research focuses on using analytics to turn data into models, and using models to facilitate decision-making. The data-models-decisions framework can be applied to a host of application areas including healthcare, supply chain management, operations of transportation agencies, and revenue management. Diwakar Gupta’s current and previous projects span the spectrum of these applications. Two ongoing projects are described below. More information can be found at the Supply Chain and Operations Research Laboratory.
Fare compliance on public transportation systems
Twin Cities-based Metro Transit uses a Proof of Payment (PoP) system with barrier-free stations on its Hiawatha Light Rail Line (HLR) and Northstar Commuter Line. Trains and station platforms are paid zones where patrons are expected to carry proof of payment at all times. Compliance is enforced by Metro Transit Police performing spot checks and issuing citations to patrons found without proof of having paid the fare. Metro Transit also performs random sampling to estimate non-compliance and ridership.
Metro Transit needs to estimate and manage fare compliance to calculate missed revenue and undertake efforts to increase revenue via better compliance. The purpose of this project is to develop a suite of methodologies for estimating compliance by utilizing data collected for other purposes (ticket sales, tagged rides, mobile validator used by Metro Transit Police, and audits), sampling, and crowd sourcing. All transit agencies that use PoP system can benefit from such methodologies. In this project, the research team is developing statistically sound methodologies for fare compliance estimation as well as inspection strategies that are most effective at reducing non-compliance.
Organ transplant waiting lists
Transplantation is often the only treatment available to patients suffering from end-stage organ disease. More than 117,000 patients are currently waiting for different organs, but only about 25,000 receive transplants in a year. The extreme shortage of organs has led to long wait times and thousands of waitlist deaths each year. This shortage also makes it necessary to prioritize matched candidates for each available organ.
Pursuant to the National Organ Transplant Act, the difficult task of setting allocation priorities rests with the Organ Procurement and Transplantation Network, run by the United Network of Organ Sharing (UNOS). The Scientific Registry of Transplant Recipients (SRTR) supports organ transplant operations by performing policy evaluation. SRTR uses Simulated Allocation Models (SAMs) to evaluate the impact of proposed allocation policies on the distribution of organs, waitlist statistics, and post-transplant outcomes. SAMs are computer programs that simulate the placement of deceased-donor organs according to specified allocation policies, and then collect a variety of statistics of interest. A key module in SAMs is a classifier that determines whether a matched candidate under a specific allocation policy will accept or decline an offered organ. The classifier needs to produce realistic results based on donor, candidate, and policy attributes that drive such decisions in practice. In this project, Gupta's team has used data mining and analytical techniques to evaluate different classifiers for making accept/decline decisions for liver candidates.
Tang, Y., Gurnani, H., and Gupta, D. 2013. Managing Disruptions in Decentralized Supply Chains with Endogenous Supply Process Reliability. Production and Operations Management, Accepted for publication
Chen, H.W., Gupta, D., and Gurnani, H., 2013. Fast-Ship Commitment Contracts in Retail Supply Chains., IIE Transactions, Vol. 45, No. 8, 811-825. DOI:10.1080/0740817X.2012.705449
Gupta, D., Li, F., and Wilson, N., 2011. Extraboard-Driver Workforce Planning for Bus Transit Operations, CURA REPORTER, Volume 41, No. 2 (Summer 2011).
Gupta, D., and Wang, W-Y., 2011. Patient Appointments in Ambulatory Care, Chapter 4 in Handbook of Healthcare System Scheduling: Delivering Care When and Where It is Needed, Randolph W. Hall, Editor, Springer, NY.
Wang, W-Y., and Gupta, D., 2011. Adaptive Appointment Systems with Patient Preferences. Manufacturing & Service Operations Management, 13 (3), 373-389.