Research Interests: capacity and patient flow management in health care operations, diffusion models for new products and services, revenue management
Professor Savin’s research expertise is centered on operational aspects of health care delivery, improving patient access to care, and optimal management of diagnostic and treatment capacity. His articles have appeared in Management Science, Operations Research, and Manufacturing and Service Operations Management, among others, and he also actively participates in editorial activities for several premier journals including Management Science, Operations Research, Manufacturing and Service Operations Management, and Production and Operations Management.
Professor Savin teaches a PhD course on optimization, the core MBA course on Business Analytics, and the core undergraduate course on Operations and Information Management.
Before joining the Wharton School in July 2009, Professor Savin was on the faculty at the Columbia Business School and the London Business School. He received a Ph.D. in Physics from the University of Pennsylvania in 1997 and a Ph.D. in Operations and Information Management from the Wharton School in 2001.
Nan Liu, Sergei Savin, Matthew Steinberg, Vanitha Virudachalam (Work In Progress), Education as a Coproduced Service: Incentivizing Teachers and Students.
Hessam Bavafa, Lerzan Ormeci, Sergei Savin, Vanitha Virudachalam (Work In Progress), Elective Admission and Patient Discharge Policies in Hospital Environments.
Vanitha Virudachalam, Sergei Savin, Matthew Steinberg (Under Revision), Investing in Performance: Information and Merit-Based Incentives in K-12 Education.
Abstract: The United States educational policy requires that K-12 students participate in annual standardized tests. As a result, school districts that have traditionally utilized ongoing “formative” assessments of student progress, are increasingly relying on additional, costly “interim” assessments. In addition, some districts are experimenting with merit-based incentives that tie teachers’ bonuses to student performance on state tests. We examine the relationship between information on student performance and monetary incentives for teachers using a two-period principal-agent model. In our model, the school district (principal) chooses whether to invest in interim assessments, and, also, how much merit-based compensation to offer to teachers, while the teachers (agents) decide on the level of effort to exert in each period. We use two-state (“proficient” vs. “not proficient”) Markovian dynamics to describe the evolution of student readiness for the tests, and assume the presence of information asymmetry between the teachers and the school district regarding the student readiness level. Our analysis shows that, for schools that are not proficient at the beginning of the year, the return from merit-based incentives is always greater than the return from information derived from interim assessments. For schools that begin the year on track to achieve proficiency, there exist settings where investing in the interim assessment is optimal, such as when the district has a low budget and the formative assessment is reasonably accurate. However, we also establish that there are settings where the provision of additional information about the student mid-year performance has a demotivating effect on teachers.
Abstract: We develop a model of crowdfunding dynamics that maximizes revenue for a given fundraising campaign by optimizing both the pledge level sought from donors or backers and the duration of the campaign. Our model aligns with the patterns of backer/donor arrival and pledging observed on crowdfunding platforms, such as Kickstarter. Using our model, we calibrate the revenue lost from using pre-specified pledge levels or campaign durations. We show that under the optimal design, the pledge level sought decreases as the goal of a campaign increases, with a more pronounced effect for both very low and very high campaign goals. We further demonstrate how uncertainty in pledge accumulation improves campaign revenue and aids campaign success. In particular, we show that campaigns with high goals benefit from highly uncertain environments more than campaigns with low goals.
Hessam Bavafa, Lerzan Ormeci, Sergei Savin (Under Review), Optimal Mix of Elective Surgical Procedures Under Stochastic Patient Length of Stay.
Linda Green, Sergei Savin, Yina Lu (2013), Primary Care Physician Shortages Could Be Eliminated Through Use of Teams, Non-Physicians, and Electronic Communication, Health Affiars, 32, pp. 11-19.
Houyuan Jiang, Zhan Pang, Sergei Savin (2012), Performance-Based Contracting for Outpatient Medical Services, Manufacturing and Service Operations Management, 14 (4), pp. 654-668.
Adam Powell, Sergei Savin, Nicos Savva (2012), Physician Workload and Hospital Reimbursement: Overworked Servers Generate Less Revenue per Patient, Manufacturing and Service Operations Management, 14 (4), pp. 512-528.
Abstract: In our prior work on product diffusions in presence of a capacity constraint, we postulated that a firm operating in such an environment should always attempt to fulfill as much of the present demand as is possible with the capacity constraint. In other words, the firm would never have demand backlogged while accumulating inventory. In this note, we derive a sufficient condition for the optimality of such fulfillment policy.
Sergei Savin, Houyuan Jiang, Serguei Netessine (2010), Robust Newsvendor Competition Under Asymmetric Information, Operations Research, Articles in Advance, pp. 1-8.
Abstract: We generalize analysis of competition among newsvendors to a setting in which competitors possess asymmetric information about future demand realizations, and this information is limited to knowledge of the support of demand distribution. In such a setting, traditional expectation-based optimization criteria are not adequate, and therefore we focus on the alternative criterion used in the robust optimization literature: the absolute regret minimization. We show existence and derive closed-form expressions for the robust optimization Nash equilibrium solution for a game with an arbitrary number of players. This solution allows us to gain insight into the nature of robust asymmetric newsvendor competition. We show that the competitive solution in the presence of information asymmetry is an intuitive extension of the robust solution for the monopolistic newsvendor problem, which allows us to distill the impact of both competition and information asymmetry. In addition, we show that, contrary to the intuition, a competing newsvendor does not necessarily benefit from having better information about its own demand distribution than its competitor has.
OIDD 101 explores a variety of common quantitative modeling problems that arise frequently in business settings, and discusses how they can be formally modeled and solved with a combination of business insight and computer-based tools. The key topics covered include capacity management, service operations, inventory control, structured decision making, constrained optimization and simulation. This course teaches how to model complex business situations and how to master tools to improve business performance. The goal is to provide a set of foundational skills useful for future coursework atWharton as well as providing an overview of problems and techniques that characterize disciplines that comprise Operations and Information Management.
"Managing the Productive Core: Business Analytics" is a course on business analytics tools and their application to management problems. Its main topics are optimization, decision making under uncertainty, and simulation. The emphasis is on business analytics tools that are widely used in diverse industries and functional areas, including operations, finance, accounting, and marketing.
The specific content of this course varies from semester to semester, depending on student and faculty interest.
This course constitutes the second part of a two-part sequence and serves as a continuation of the summer math camp. Mathematical optimization provides a unifying framework for studying issues of rational decision-making, optimal design, effective resource allocation and economic efficiency. It is a central methodology of many business-related disciplines, including operations research, marketing, accounting, economics, game theory and finance. In many of the disciplines, a solid background in optimization theory is essential for doing research. This course provides a rigorous introduction to the fundamental theory of optimization. It examines optimization theory in two primary settings: static optimization and optimization over time (dynamic programming). Applications from problem areas in which optimization plays a key role are also introduced. The goal of the course is to provide students with a foundation sufficient to use basic optimization in their own research work and/or to pursue more specialized studies involving optimization theory. The course is designed for entering doctoral students. The prerequisites are calculus, linear algebra and some familiarity with real analysis, as covered in summer math camp. Other concepts are developed as needed throughout the course.
This calculus-based course is recommended for science majors and engineering students. Classical laws of motion; interactions between particles; conservation laws and symmetry principles; particle and rigid body motion; gravitation, harmonic motion, and applications of mechanics to real-world problems. Credit is awarded for only one of the following courses: PHYS 008, PHYS 101, 150, 170. Students with AP or Transfer Credit for PHYS 91 or 93 who complete PHYS 150 will thereby surrender the AP or Transfer Credit. Prerequisite: Students in PHYS 150 should already have taken MATH 104 or the equivalent, or be taking it simultaneously with PHYS 150.
The topics of this calculus-based course are electric and magnetic fields; Coulomb's, Gauss's, Ampere's, and Faraday's laws; DC and AC circuits; Maxwell's equations and electromagnetic radiation. Credit is awarded for only one of the following courses. PHYS 009, 102, 151, 171. Students with AP or Transfer Credit for PHYS 92 or 94 who complete PHYS 151 will thereby surrender the AP or Transfer Credit. Prerequisite: Students in PHYS 151 should already have taken MATH 114 or the equivalent, or betaking it imultaneously with PHYS 151.