Faculty Details

Photo of Rema  Padman

Rema Padman

Professor of Management Science & Healthcare Informatics

Email: rpadman@andrew.cmu.edu


Professor Padman's research addresses problems at the interface of healthcare, information technology and management science, particularly healthcare information systems, operational planning and management, and data mining and decision support methods. Her current research in the healthcare domain investigates data mining methods for healthcare decision support; evaluating the use and impact of information technology and systems in healthcare environments, particularly for point-of-care disease management; and, examining tradeoffs between access and confidentiality in large multidimensional public-use and healthcare databases. Her research on these topics has been funded by the National Science Foundation, National Library of Medicine, DARPA, and the Army Research Office.

Prior to joining the Heinz School's faculty in 1989, Professor Padman was Assistant Professor of Management Science in the Carlson School of Management at the University of Minnesota. More recently, she was Visiting Scientist at Humboldt University in Berlin, Germany. She is an Associate Editor with the INFORMS Journal on Computing and a past Associate Editor with Operations Research. She has also served on the National Science Foundation (NSF) panels on Operations Research grants and on several program committees of information systems and operations research conferences.

Professor Padman has received the Emil Limbach Award for Excellence in Teaching and the Outstanding Instructor Award in the Woodrow Wilson Summer Studies program, both at the Heinz School at Carnegie Mellon University. She is also a recipient of the Individual National Research Service Award from the National Library of Medicine, a division of the National Institutes of Health. Professor Padman's paper titled "On Analyzing Interactions in a Software Agent Marketplace" (with Heinz School colleagues) was awarded the best paper award at 2000 Workshop on Information Technologies and Systems in Brisbane, Australia.



"Brokering Decision Support Resources for Supply Chain Management" (with A. Asavanund, H. Bhargava, D. Fernandes, and R. Krishnan), Agents '99 Workshop on Supply Chain Management, May 1999, forthcoming.

"The Diffusion of Information Technology Among Health Maintenance Organizations" (with D.R. Wholey, R. Hamer, S. Schwartz, and R. Stillman), Health Care Management Review, forthcoming.

"On Payment Scheduling in Client-Contractor Negotiations: An Overview of the Problem and Research Issues" (with N. Dayanand), Project Scheduling: Recent Models, Algorithms and Applications (J. Weglarz, ed.), in the series Handbook of Operations Research (F. Hillier, ed.), pp. 477-508.

"On Intelligent Brokering of Web Based Computational Services" (with A. Asvanund, D.F. Fernandes, and R. Krishnan), Proceedings of the INFORMS Conference on Information Systems and Technology, Montreal, Canada, 1998.

"Accounting Information System Data Quality Assessment: A DSS Approach" (with D. Kaplan, R. Krishnan, and J. Peters), Communications of the ACM, Vol. 41, NO. 2, pp. 72-78, 1998.

"Disclosure Detection in Multiple Linked Categorical Datafiles: A Unified network Approach" (withS. Roehrig, G. Duncan, and R. Krishnan), Proceedings of Statistical Data Protection Conference '98, 1998.

"Inference of Three-Way Table Entries from Two-Dimensional Projections" (with G. Duncan, R. Krishnan, and S. Roehrig), Proceedings of HICSS '98, 1998.

"Quality Metrics for Healthcare Data: An Analytical Approach" (with M. Tzourakis), Proceedings of MIT IQ '97, 1997.

"On Modeling Payments in Projects" (with N. Dayanand), Journal of the Operational Research Society, Vol. 48, pp. 906-918, 1997.

"Efficient Distributed Simulation through Load Balancing" (with Murali Shanker and W. David Kelton); IIE Transactions, 1997.

"Connectionist Approaches for Solver Selection in Constrained Project Scheduling" (with Dan Zhu); Annals of Operations Research, Vol. 72, pp. 265-298, 1997.

"Heuristic Scheduling of Resource-Constrained Projects with Cash Flows" (with Dwight Smith-Daniels and Vicki Smith-Daniels); Naval Research Logistics, Vol. 44, pp. 364-381, 1997.

"A Genetic Programming Approach for Heuristic Selection in Constrained Project Scheduling" (with Stephen Roehrig); Interfaces in Computer Science and Operations Research: Advances in Metaheuristics, Optimization, and Stochastic Modeling Technologies, (R.S. Barr, R.V. Helgason, and J.L. Kennington, eds.), Kluwer Academic Publishers, Norwell, MA, pp. 405-421, 1997.

"A Cooperative Multi-Agent Approach to Constrained Project Scheduling" (with Dan Zhu); Interfaces in Computer Science and Operations Research: Advances in Metaheuristics, Optimization, and Stochastic Modeling Technologies, (R.S. Barr, R.V. Helgason, and J.L. Kennington, eds.), Kluwer Academic Publishers, Norwell, MA, pp. 367-381, 1997.

"Cell Suppression to Limit Content-based Disclosure" (with George Duncan, Ramayya Krishnan, Phyllis Reuther, and Stephen Roehrig); Proceedings of HICSS Conference, 1997.

"On Using Web Technologies to Architect DSS: The Case of Support Requirements Planning" (with Ramayya Krishnan); Proceedings of the ISDSS Conference, 1997.

"Heuristic Scheduling of Capital Constrained Projects" (with Dwight Smith-Daniels and Vicki Smith-Daniels); Journal of Operations Management, Vol. 14, pp. 241-254, 1996.

"A DSS Approach to Information System Reliability Assessment" (with David Kaplan, Ramayya Krishnan, and James Peters); Proceedings of the INFORMS Conference on Information Systems and Technology, Washington, DC, 1996.

"Intelligent Decision Support for Constrained Project Scheduling" (with Dan Zhu); Proceedings of the INFORMS Conference on Information Systems and Technology, Washington DC, 1996.


PhD, Operations Research, University of Texas at Austin

Working Papers

Tabu Search Enhanced Markov Blanket Classifier for High Dimensional Data Sets

Data sets with many discrete variables and relatively few cases arise in health care, ecommerce, national security, and many other domains. Learning effective and efficient prediction models from such data sets is a challenging task. This paper proposes a Tabu Search enhanced Markov Blanket (TS/MB) procedure to learn a graphical Markov Blanket classifier from data. The TS/MB procedure is based on the use of restricted neighborhoods in a general Bayesian network constrained by the Markov condition, called Markov Equivalent Neighborhoods. Computational results from real world data sets drawn from health care domain indicate that the TS/MB procedure converges fast, is able to find a parsimonious model with substantially fewer predictor variables than in the full data sets, gives comparable or better prediction performance when compared against several machine learning methods, and provides insight into possible causal relations among the variables.


Optimal Bidding in Sequential Online Auctions

Auctions are widely used online to conduct commercial transactions. An important feature of online auctions is that even bidders who intend to buy a single object frequently have the opportunity to bid in sequential auctions selling identical objects. This paper studies key features of the optimal bidding strategy, assuming rational, risk-neutral agents with independent private valuations and sealed-bid second-price sequential auctions. In contrast to previous work on this topic, we develop our theory using the concept of the "option value" of an upcoming auction - a measure of the expected payoff from being able to participate in a future auction. This option value depends, among other things, upon the mean and variance of the future number of bidders. This paper shows an optimal bidding strategy in sequential auctions that incorporates option value assessment. Furthermore, it is establshed that that optimal bidding strategy is tractable since it is independent of the bidding strategies of other bidders in the current auction and is only dependent on the option value assessment.


User Acceptance and Adoption of a Clinical Reminder System in Ambulatory Care: A Developmental Trajectory Approach

Evaluation studies of clinical decision support systems (CDSS) have tended to focus on assessment of system quality and clinical performance in a laboratory setting. Relatively few studies have used field trials to determine if CDSSs are likely to be used in routine clinical settings and whether reminders generated are likely to be evaluated by end-users. This paper argues that such beneficial outcomes are not likely to occur if use of the system results in side-effects such as decreased end-user efficiency and unanticipated changes in normal workflows.


Clinical Reminder System: A Relational Database Application for Evidence-Based Medicine Practice

Evidence-based medicine is the distillation of a large volume of medical research and standards into treatment protocols for diseases and preventive care procedures that represent the most accurate knowledge available. In this project, we implement evidence-based medicine principles via a decision support system that provides suggested actions for physicians based on individual patient characteristics and established treatment protocols. Such a reminder system may enable physicians to make better-quality decisions, and may enable patients more consistently follow medical recommendations. This papers presents a prototype DSS, called Clinical Reminder System, that combines a relational database, a knowledge base consisting of algorithms that implement disease treatment protocols, integration with hospital legacy systems and a web-based interface allowing for physician management of patient data and suggested medical responses. This application has been in use within a clinical setting since 2001. Formal evaluation and assessment of patient outcomes associated with use of this system is currently being performed by Carnegie Mellon University and The Western Pennsylvania Hospital.



Determinants of Information Technology Outsourcing Among Health Maintenance Organizations

This paper extends transaction cost economics by examining the effect of relaxing two of its underlying assumptions. First, transaction cost economics relies on an assumption of risk neutrality. This paper argues that organizations transactions vary in the risk they impose on an organization and that organizations are more likely to embed riskier transactions within a hierarchy. Second, transaction cost economics assumes that transactions are independently organized. Organizations have an underlying propensity to organize transactions through hierarchy or contracting and that this underlying propensity is related to an organization’s capabilities, such as absorptive capacity. The analysis shows that transaction organization is a function of transaction risk. Transaction risk, rather than uncertainty or firm asset specificity, is the most important factor determining transaction organization. And, the analysis shows that transaction organization is a function of an organization’s absorptive capacity and technological diversity. This means that transactions within an organization are interdependent.