Mark Lewis

Born:

place:

Bachelor of Science & Arts Degree (1992) Mathematics & Political Science Eckerd College, St. Petersburg, FL; MS (1995) (Theoretical Statistics), Florida State University

Ph.D. 1998, Georgia Institute of Technology, (Industrial and Systems Engineering)
thesis: ; Advisors: Hayriye Ayhan and Robert D. Foley

area of degree: Stochastic Dynamic Programming

Assistant Professor University of Michigan

personal or universal URL: http://www-personal.engin.umich.edu/~melewis or http://ioe.engin.umich.edu/people/fac/melewis.html

email: melewis@umich.edu

 

Professor Lewis was hired as an Assistant Professor by the University of Michigan in 1999. His research interests include decision-making under uncertainty when the objective is a non-standard objective function. This leads to extensions of basic ideas in Markov decision processes and applied probability. Applications of such research include optimal admission, pricing, and control of queueing and inventory systems.

ACADEMIC HONORS

• NSF CAREER Grant DMI-0132811  (Principal Investigator ) -- Monotone Optimal Policies in Parallel Processing Networks (2002-present)
• NSF Grant DMI-9908321 (Principal Investigator ) -- Optimal Admission and Pricing in Non-stationary Queueing Systems (1999-present)
• Georgia Tech FACES Research Award (1st ever recipient)  (1999)
• Runner-up INFORMS George B. Dantzig Dissertation Award (1999)
• NSF-NATO Postdoctoral Fellow, 1998-1999

Research

Dr. Lewis' interests are in Stochastic Processes with an emphasis on queueing theory. He explained implicit discounting in bias optimality and again related it to controlled queueing systems. Most recently, he is interested in parallel processing and how resources are allocated dynamically in such systems.

PUBLICATIONS

• Hyun-soo Ahn, Izak Duenyas, and Mark E. Lewis, "The Optimal Control of a Two-Stage Tandem Queueing System with Flexible Servers", to appear in Probability in the Engineering and Informational Sciences, Vol. 4 (2002).
• Jason H. Goto, Mark E. Lewis, and Martin L. Puterman, "Coffee, Tea, or ...?: A Markov Decision Process Model for Airline Meal Provisioning", 2002. To appear in Transportation Science.

1. Cooper, William L.; Henderson, Shane G.; Lewis, Mark E. Convergence of simulation-based policy iteration. Probab. Engrg. Inform. Sci. 17 (2003), no. 2, 213--234.
2. Ahn, Hyun-soo; Duenyas, Izak; Lewis, Mark E. Optimal control of a two-stage tandem queuing system with flexible servers. Probab. Engrg. Inform. Sci. 16 (2002), no. 4, 453--469.
3. Lewis, Mark E.; Ayhan, Hayriye; Foley, Robert D. Bias optimal admission control policies for a multiclass nonstationary queueing system. J. Appl. Probab. 39 (2002), no. 1, 20--37.
4. Lewis, Mark E.; Puterman, Martin L. Bias optimality. Handbook of Markov decision processes, 89--111, Internat. Ser. Oper. Res. Management Sci., 40, Kluwer Acad. Publ., Boston, MA, 2002.
5. Lewis, Mark E. Average optimal policies in a controlled queueing system with dual admission control. J. Appl. Probab. 38 (2001), no. 2, 369--385.
6. Lewis, Mark E.; Puterman, Martin L. A probabilistic analysis of bias optimality in unichain Markov decision processes. IEEE Trans. Automat. Control 46 (2001), no. 1, 96--100.
7. Lewis, Mark E.; Puterman, Martin L. A note on bias optimality in controlled queueing systems. J. Appl. Probab. 37 (2000), no. 1, 300--305.
8. Lewis, Mark E.; Ayhan, Hayriye; Foley, Robert D. Bias optimality in a queue with admission control. Probab. Engrg. Inform. Sci. 13 (1999), no. 3, 309--327.

a young Lewis

references: Mark Lewis and his web page; Mathematical Reviews

The web pages
MATHEMATICIANS OF THE AFRICAN DIASPORA
are brought to you by

The Mathematics Department of
The State University of New York at Buffalo.

They are created and maintained by
Scott W. Williams
Professor of Mathematics

CONTACT Dr. Williams