Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

The theoretical discussion covers the performance of descent algorithms in terms of worst-case analysis and the time complexity of searching for a local optimum. Also, we discuss some methods that allow neighborhoods of exponential size to be searched in polynomial time.

The last part of the talk focuses on some of the more recent types of local search methods. Specifically, the potential for ant colony optimization, variable neighborhood search, multi-level local search, and other methods to produce competitive local search algorithms for scheduling problems is discussed.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

portfolio several components are necessary. We will take a look at the components that was used to implement OptRet. Topics that we will cover are choice of objective function, optimization model, modeling language, solver, interface and preliminary results. A brief overview of the interior point solver and its properties will be given. We will discuss two versions of a new modeling language. The earlier version

has syntax similar to AMPL, with extensions to stochastic programming, and the later version will be a general open source modeling language with similar flexibility as AMPL.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Each year, the ISyE/Math Center for Applied Probability at

Georgia Tech hosts a two-day Southeast Probability Days Conference.

This year, our neighbors from the University of Tennessee have invited us to combine our conference with their probability conference under the title of Barrett Memorial Lectures. See the details below. We encourage you to attend.

----------------------------------

Dear Colleague:

We are organizing the "John H. Barrett Memorial Lectures" that will be held at the University of Tennessee in April 25 - 26, 2003. The tradition of these Lectures began in 1970 to honor our distinguished scientist John H. Barrett. The Lectures have been given annually since 1972 by a succession of distinguished speakers in various areas of research interest of our mathematics faculty.

This year probability has been chosen as the area for the Lectures Series.

"Random walks, Levy processes, and related topics" will be the general theme.

Jean Bertoin (Paris VI) and Greg Lawler (Cornell Univ.) are the principal speakers of the Series; each will deliver three one-hour talks.

We have also planned to have five additional half-hour lectures in the realm of the theme mentioned above. The details of the academic program and other related activities are included in conference website:

http://www.math.utk.edu/Barrett/

We invite you to participate in the Lecture Series. We have applied for some funds to partially support those participants who do not have other means of support. Should the funds become available, preference will be given to young researchers and graduate students. Should you accept this invitation, please fill the electronic registration form included in the conference website by March 24, 2003. If you know any graduate students who you think may benefit by participating in this meeting, please encourage them to apply.

We have reserved a block of rooms for the participants in Days Inn and Hilton Hotel which will be held until April 1. As you may know, April is a beautiful time to visit Knoxville and the near by Great Smoky Mountains. Many activities are taking place here at this time. We therefore urge you please make your own reservations as soon as possible. Hotel's addresses and telephone numbers are provided on the conference website.

Further inquires regarding the scientific program can be obtained from the organizers listed below and, regarding the support and accommodation, from Ms. Alice Gregory gregory@math.utk.edu

Sincerely,

Jan Rosinski

(for the Organizing Committee)

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

the press and regulators, in that order. We introduce and define hedge funds, dispel some commonly held misconceptions, explain why they have outperformed especially in a difficult market environment, discuss their characteristics and strategies, present an overview of the hedge fund industry today, and introduce funds of hedge funds. We will focus on roles and strategies that are particularly geared to individuals with engineering

backgrounds.

Speaker:

Dr. Jason Papastavrou

Funds of Hedge Fund Strategies

Banc of America Capital Management

40 West 57th Street (33rd Floor)

New York, NY 10019

Dr. Papastavrou is responsible for portfolio management of all Fund of Hedge Funds products. Prior to joining BACAP in 2001, he was a portfolio manager for Deutsche Asset Management from 1999 to 2001. Prior to that he worked for a private investor and was a professor at Purdue University from 1990 to 1999. Dr. Papastavrou has been a member of the investment community since 1997. Dr. Papastavrou earned a Ph.D. and master's degree in electrical engineering and a bachelor's degree in mathematics all from the Massachusetts Institute of Technology. He is a member of The Institute for Operations Research and Management Science and The Institute of Electrical and Electronic Engineers.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

To maintain expertise in the face of rapid change, ongoing training of expert personnel is a necessity. It is difficult, however, with current training methods, e.g., classroom training, to ensure that practitioners keep abreast of new knowledge and procedures. Training expense and logistics are major obstacles.

The goals of this research are to address the growing training demands of maintaining practitioner expertise by using computer-based training that merges ITS and case-based teaching. A further goal is to implement this new approach in such a way that facilitates the ease and decreases the cost of incorporating new cases as training needs evolve. To address these goals, this research proposes a theory and architecture for case-based computer-implemented training, Case-Based Intelligent Tutoring System (CBITS). CBITS builds upon the experience and research in both ITS and case-based teaching. The ITS provides a control structure for monitoring the individual student and addressing their individual needs. Within that structure, cases provide a method of teaching, using memorable experiences to create focused instruction. Cases also allow tutor content to evolve as the operational environment evolves.

Implementations of CBITS are relevant in a range of domains in which practitioners interact with a technological and evolving system. Examples include airline pilots and maintenance, electronic manufacturing, and telecommunications. Currently, CBITS is implemented in proof-of-concept for MD-11 pilots, teaching a newly licensed capability of the aircraft. This capability introduces a new technique that improves safety but is unfamiliar to experienced pilots. An evaluation of the system with active airline pilots showed the system and training to be highly effective.

Committee

----------------------------------------------------------------------

Prof. Christine M. Mitchell (Advisor)

Prof. T. Govindaraj

Prof. Dave Goldsman

Prof. William B. Rouse

Dr. Everett A. Palmer, III (NASA Ames Research Center)

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

1) Buffer space allocation problem for assembly/disassembly production systems;

2) Kanban and initial inventory allocation problem for extended kanban control systems;

3) Service capacity allocation problem for assembly/disassembly production systems;

4) Release time determination problem for tandem line production systems;

All production systems considered here are modeled by Fork/Join type queueing network systems with blocking, and the performance measures to be optimized are manufacturing efficiencies such as throughput, lead time and work-in-process.

In general, in order to optimize queueing network systems with blocking, we have to resolve two difficult tasks: one is to evaluate the values of the performance measures, and the other is to search the optimal design parameters. For the first task (calculating values of performance measures), we could devise approximate Markov analysis or run simulations. Here, we put focus on the simulation-based approaches. Having an approximation scheme for calculating the performance measure established, we could use conventional optimization methods to optimize it. We apply total enumeration or meta-heuristics such as Genetic Algorithm for discrete optimal design problems, and non-linear optimization methods for continuous problems. In particular, the service capacity allocation problem is formulated as a Second Order Cone Programming problem (SOCP), which can be solved effectively. While, the release time determination problem results in a global minimization of difference piecewise linear convex functions which seems to be difficult to solve.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Andre Rohe from Sun Microsytems will give a talk on the algorithmic problems arising in the physical design of microprocssors. Andre has his PhD from the Institute for Discrete Mathematics in Bonn, Germany, and he is now part of the processor design team at Sun Microsystems in Sunnyvale, CA.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

We then analyze the asymptotic tail distribution of stationary waiting times and stationary virtual waiting times in a single server queue with long-range dependent arrival process and subexponential service times. We investigate the joint impact of the long range dependency of the arrival process and of the tail distribution of the service times. We consider two traffic models that have been widely used to characterize the long-range dependence structure, namely, the fractional Gaussian noise (FGN) model and the M/G/infty input model. We show that the asymptotic tail distribution of the waiting time is dominated by either the arrival process or the service times, depending on how large is the Hurst parameter of the

arrival process compared to the service time distribution.

This is joint work with Cathy Xia at IBM Research.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Venue: Rock Bottom Restaurant & Brewery (503) 796-2739

210 SW Morrison

Portland, OR 97204-3109

About 5 blocks west of the conference hotel. RSVP not required. Just show up and catch up.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Eric Bonabeau is the chief scientist at Icosystem Corporation, a Cambridge, MA-based "idea incubator" that uses complexity science to invent new technologies. Prior to his current position, Dr. Bonabeau was the CEO of Eurobios, a joint venture with Cap Gemini Ernst & Young applying the science of complex adaptive systems to business issues. He has been a research engineer with France Telecom R&D, an R&D engineer with Cadence Design Systems, and the Interval Research Fellow at the Santa Fe Institute. He is the author of more than one hundred science articles and three books (Intelligence Collective, Hermhs, 1994; Swarm Intelligence in Natural and Artificial Systems, Oxford University Press, 1999; and Self-Organization in Biological Systems, Princeton University Press, 2001). Dr. Bonabeau is also co-editor-in-chief of Advances in Complex Systems and a member of the editorial and scientific committees of more than twenty-five international journals and conferences. He graduated from Ecole Polytechnique, France, holds a telecommunications engineering degree from Telecom Paris, a post-graduate degree in applied mathematics and a PhD in theoretical physics both from Paris X University.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Your student INFORMS chapter is pleased to announce its first seminar of the summer.

Mr. Glenn Bailey of Delta Airlines will be discussing "Organization, Engagement Aspects, and Current Types of Projects

in Operations Research at Delta Airlines"

Pizza and drinks will be provided. We hope all of you can come!

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

The talk summarizes an analysis of a unique record of call center operations. The data comprise a complete operational history of a small banking call center, call by call, over a full year. Taking the perspective of queueing theory, we decompose the service process into three fundamental components: arrivals, waiting times, and service durations. Each component involves different basic mathematical structures and requires a different style of statistical analysis. Some of the key results will be sketched, along with descriptions of the varied techniques required.

In conclusion we survey how the characteristics deduced from the statistical analyses form the building blocks for theoretically interesting and practically useful mathematical models for call center operations.

This reports on joint work with Larry Brown, Linda Zhao and Noah Gans from Wharton, and Avishai Mandelbaum, Anat Sakov and Sergey Zeltyn from Technion (Israel).

]]>Industrial and Systems Engineering

Contact Barbara Christopher

The second part of this thesis is a study of a

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

The main objective of this dissertation is to accomplish the fusion by answering the question "What is the relationship between the efficiency scores provided by DEA and the partial productivity metrics?" We first connect TE and PE assuming a priori costs/prices information. We show PE is a special case of bilateral comparison of the performance of two organizations. Integrating the early works, the bilateral comparison, thus, can be further decomposed into detail comparisons based on eleven metrics. We also build the connection between TE and PE directly. We show that both PE and TE can be computed using similar LP formulation. Therefore, a sequence of LP models, which starts from TE and ends at PE, collectively, forms a bridge between PE and TE. This bridge has several "spans", each corresponding to a particular effect. Therefore, a particular PE can be decomposed into seven multiplicative factors including TE. This theoretical linkage provides aids for output-input ratio benchmarking performance gap analysis and leads to some practical guidelines for selecting a PE to approximate system-based efficiency when DEA is not a possible solution. In the end, we present a warehouse benchmarking study.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

======================================================================================================

Please plan to attend what promises to be an interesting, perhaps even provocative talk.

Professor Hodgson is an IE faculty member and former department head at NC State, retired sky-diver, and recent inductee to the National Academy of Engineering.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Detection) have been suggested when multiple TCP sessions are multiplexed through a bottleneck buffer. The idea is to detect congestion before the buffer overflows and packets are lost. When the queue length reaches a certain threshold RED schemes drop/mark incoming packets with a probability that increases as the queue size increases. The objectives are an equitable distribution of packet loss, reduced delay and delay variation and improved network utilization. Here we model multiple connections maintained in the congestion avoidance

regime by the RED mechanism. The window sizes of each TCP session evolve like independent dynamical systems coupled by the queue length at the buffer. We introduce a mean-field approximation to one such RED system as the number of flows tends to infinity. The deterministic limiting system is described by a transport equation. The numerical solution of the limiting system is found to provide a good description of the evolution of the distribution of the window sizes, the average queue size, the average loss rate per connection and the total throughput.

He will be giving a follow-on seminar in the Probability Seminar Series:

Thursday Feb 20 at 3 PM in Room 269

Title: Mean-field convergence of distributed dynamical systems sharing a common resource

Abstract: Multiple TCP/IP connections multiplexed through a RED buffer are in fact dynamical systems which share a common resource - the buffer. We adapt the technique developed by Kurtz and Donnelly to prove mean-field convergence of the histogram of window sizes.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

things from data directly, but the issue is what to plot with a

high-dimensional data set. Thus dimension reduction shows its

importance. Based on information theory, here we develop a model-free dimension reduction technique for extracting important information. Our method in general could be applied in any area relating to information extraction. It also can be viewed as inverse regression. It involves density estimation which can be estimated non-parametrically. And that is feasible with the help of fast computing techniques. We provide theoretical justification for connecting with central

subspace. Illustrative examples are presented.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

process are candidates for such asymptotic processes. Examples for these limiting situations are given and the limiting distribution of the minimum value and the minimizer are discussed. It turns out, that asymptotic normality appears only as an important, but special case, other types of limiting distributions appear in a natural way.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Want to learn more about what IE's do in their jobs?

A FREE lunch (Papa John's pizza) will be provided- first come first serve!

For more information on how you can be involved in your professional society, contact:

Melissa Krenzel

IIE President 02-03

Industrial and Systems Engineering

Contact Barbara Christopher

vaccine strains recommendation to countries around the globe. However,

various studies have found that the WHO vaccine selection strategy has

not been effective in some years. This motivates the search for a better

strategy for choosing vaccine strains. In this work, we use recent results

from theoretical immunology to formulate the vaccine selection problem as a

discrete-time stochastic dynamic program with a high-dimensional

continuous state space. We discuss the techniques that were developed

for solving this difficult dynamic program, and present an effective

heuristic policy. We also compare the performance and practicability of

the optimal policy, the WHO policy, and the heuristic policy.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

discussed. It is high reliability, not an efficient scheduling, that can enhance yield of microelectronics products. Stress burn-in and issues related to burn-in, such as optimization and statistical data analysis are

introduced in this talk. Because low yield will likely be a biggest issue in nano manufacturing, this talk will also help deal with reliability enhance strategies for design and manufacturing of nano products.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

- What is risk management? Integration of real options and proprietary trading.

- Scarcity of data and need for robust modeling

- The proper use of forward-looking and historical information

- The proper use of price and non-price information

- Models as tools vs. models as "the truth": model what you can, and not what you want

- Real options and tradable risks

- Hedging of real options: static and dynamic decomposition in terms of tradables.

- Correlation and dependence modeling

- Physical assets: interaction of dispatch, hedging and valuation

Short Bio:

Krzysztof Wolyniec is the Director of Research at Mirant Corporation. He is responsible for modeling power and fuel markets as well as developing hedging and trading strategies around physical power and fuel assets. He holds graduate degrees in Physics and Operations Research from the University of Gdansk and the University of Rochester.

Industrial and Systems Engineering

Contact Barbara Christopher

Reserve a space at the URL below.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

The special problem of optimal jobs allocation to heterogeneous servers also arises in many applications including recequensing of queues. The problem of optimal jobs allocation to two heterogeneous servers with respect to the long ran average mean number of jobs in the system minimization was considered in [1], where it was shown that in this case an optimal policy has a threshold property and consists in using the fastest server if necessary. For the multi-server system these properties of optimal policies were generalized in [2].

For such a system engaged by additional cost structure those type of results are unavailable. In the talk some results of numerical analysis of such a system with and without additional cost structure will be given. This gives also the possibility to investigate the qualitative properties of optimal policies and their behavior when parameters are varied. For some numerical examples we calculate appropriate threshold levels for different values of system parameters.

References

1. W. Lin, P.R. Kumar. Optimal control of a queueing system with two heterogeneous servers. IEEE Trans. on Autom. Control, 29 (1984), pp. 696-703.

2. V.V. Rykov. Monotone Control of Queueing Systems with Heterogeneous Servers. QUESTA, 37 (2001), 391-403.

Industrial and Systems Engineering

Contact Barbara Christopher

Dr. Sogomonian will be chatting with MS QCF students after the seminar from 3pm to 4pm.

Speaker Bio:

Dr. Aram Sogomonian is Vice President, Risk Management for Constellation Power Source, the Trading and Marketing business of Constellation Energy Group. His responsibilities include taking a lead role in the risk management activities for the Trading and Marketing business and working with the CRO to design an Enterprise Wide risk organization. Prior to Constellation, Aram held a variety of management positions at Pacificorp, Edison International and Enron. Dr. Sogomonian has a Ph.D in Management Science from the Anderson Graduate School of Management at UCLA, a master of science degree in Operations Research and bachelor of arts degrees in Applied Mathematics and Economics from the University of California at Berkeley.

Host: Prof. Shijie Deng

]]>Industrial and Systems Engineering

Contact Barbara Christopher

smoothing estimators which involves only one bandwidth and does not have boundary effect. Moreover, empirical likelihood based confidence interval is obtained.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

optimization problem when the data of the problem is not fixed,

but belongs to a well-defined uncertainty set. In this scheme,

one typically aims for a solution that minimizes (or maximizes)

an objective function against all possible realizations of the

data. From a complexity point of view a desirable property of

robust optimization models is that if the nominal problem (the

one with fixed data) is solvable in polynomial time, then so is

the robust counterpart. Nemirovski et al. have introduced several

robust convex optimization models for which this property holds.

Recently Bertsimas and Sim gave an MIP model for robust discrete

optimization. They showed that when uncertainty is in the objective coefficients, if a nominal 0-1 problem is solvable in polynomial time, so is the robust counterpart. However, the given robust model has typically very weak LP bound, which makes it difficult to solve in general.

In this talk we will describe alternative models for robust 0-1

programming. In particular, we will give three models, all of

which have the strongest possible LP relaxation independent of

the nominal 0-1 problem. In addition, we will show that there

is an LP formulation for a robust 0-1 problem, polynomial in the

size of the LP formulation of the nominal 0-1 problem. We will

give extensions to robust mixed 0-1 programming and conclude

with preliminary computational experience.

Industrial and Systems Engineering

Contact Barbara Christopher

The objective of this dissertation is to explore the complexity, algorithms and applications of the Inverse Shortest Path Length Problem (ISPL). Prior researchers showed that ISPL is NP-complete. We explore the complexity of ISPL for special cases. We also introduce some heuristic algorithms and analyze their worst case performance. We test the algorithms on a set of randomly-generated instances, and observe that te results are usually within 3% of optimality. Finally, we present an application of ISPL to telecommunications bandwidth pricing, and test our heuristics on real-world data.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

We show that the end-to-end delay of the tagged flow in a tandem queueing network, and more generally in a tree network, is completely dominated by one of the queues. The dominant queue is the one with the maximal Hurst parameter. If several queues have the same maximal Hurst parameter, then we have to compare the ratio 1/s * (1- r)^H to determine the dominant queue, where r is the load of the queue.

In the case that the tagged flow is controlled through a window based congestion control mechanism, the end-to-end delay is still asymptotically Weibullian with the same shape parameter. We provide upper and lower bounds on the constant that determines the scale parameter of the corresponding Weibull distribution.

This is joint work with Marc LeLarge and Zhen Liu.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Biography:

Samer Takriti manages the Stochastic Analysis group in the Mathematical Sciences Department of IBM Research. Before joining IBM in 1996, Dr. Takriti worked on developing stochastic models for the unit commitment problem -- a large-scale mixed-integer program that arises when scheduling the generating units of an electric utility. He continued working on this subject while at IBM Research. Dr. Takriti left IBM in 1999 to become Director of Research at Enron Corporation in Houston. During his stay at Enron, he was responsible for providing mathematical and technical support for Enron Broadband Services. He also worked on a variety of problems related to electric power and water resources. Samer returned to IBM in 2000 where he continued working on issues related to deregulated power markets, such as contract evaluation and resource scheduling. He manages a group investigating the use of statistical methods, queueing theory, and stochastic optimization in the areas of pricing, workforce scheduling, manufacturing, and customer targeting. Dr. Takriti holds a Ph.D. degree in Industrial and Operations Engineering from the University of Michigan.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Kurt Palmer is an Assistant Professor in the Daniel J. Epstein Department of Industrial and Systems Engineering at the University of Southern California. His research focus is on data collection planning and empirical model building methods. He is currently developing methods intended to improve the efficiency of engineering design studies conducted on design simulators. Prior to joining the faculty at USC, Dr. Palmer was employed for six years as a Manufacturing Engineer at the Eastman Kodak Company; and for four years, he was the Owner of Quality Management Consulting in Rochester, NY. His Ph.D. in Industrial and Systems Engineering is from the Georgia Institute of Technology.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

with Equilibrium Constraints (MPECs). The approach makes use of a sequence of relaxed MPECs parameterized by a relaxation parameter vector and only performs

one log-barrier Newton step for each relaxed MPEC. Unlike previous approaches, the barrier and relaxation parameters are updated in such a way that the strict

interior of the relaxed MPEC remains nonempty even in the limit. We analyze the

convergence properties of the proposed algorithm. ]]>

Industrial and Systems Engineering

Contact Barbara Christopher

optimization problems. The concept of the universal barrier function (u.b.f) is introduced.

The key result of Yu. Nesterov and A.Nemirovsky allowing to obtain excellent complexity estimates is explained. Finally, we describe new results related to computations of u.b. fs for a very broad class of cones generated by Chebyshev systems. Connections with classical work of M. Krein , A. Nudelman and I. Schoenberg are explained. New results allow one to substantially extend the

domain of applicability of modern interior-point algorithms. ]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Then a new adaptive self-regular predictor-corrector algorithm is presented where the corrector step is defined by our self-regular proximity and the predictor step is either a self-regular or an affine-scaling step. Polynomial worst case complexity [O(sqrt(n) log(n) L) in the best case] and asymptotic quadratic convergence rate is established.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Statistics Year Book, Chinese Energy Balance and so on.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

present an interesting application to the (robust) resistor network

topology design (RNTD) problem, where the goal is to design an electrical network containing resistors only such that the dissipation is minimal, given the

external current values at the nodes of the network and assuming that the conductance values satisfy some normalizing constraint.

We present a linear model for the single-current case and

semidefinite models for the multi-current and the robust RNTD

problem. All models are illustrated by examples. It is also shown that by using duality the size of some of these models can be reduced significantly.

Industrial and Systems Engineering

Contact Barbara Christopher

It is my pleasure to formally invite you to our First Annual ISyE Ph.D. Tailgate Party! This event will take place prior to the Georgia Tech/Maryland football game on Thursday, October 23, from 3:30 - 6:30 P.M. on the lawn behind the Instructional Center.

The purpose of this event is to begin developing a sense of community among the students, faculty, and staff within the ISyE Ph.D. program. Food will be served, but ALCOHOL WILL BE PROHIBITED. There will be games and activities to promote interaction among everyone. There is no charge for ISyE faculty, staff, and Ph.D. students; however, anyone who would like to bring family members or close friends can do so for a charge of $2.00 per person.

If you would like to attend, please RSVP with the following information no later than Friday, October 17:

Name:

Characterization (i.e. Faculty, Staff, or Student):

Area of Concentration (e.g. Optimization, Stochastics, EDA): Number of Years in the Program:

Nationality:

Advisor's Name (if Student):

Students' Names (if Faculty):

Number of Guests:

Guest(s) Names:

Relation to you:

Please keep in mind that the purpose of this event is to promote community within the ISyE Ph.D. program, so we are asking that you restrict your guests to people closest to you (i.e. spouses, children, fiancee's, etc.)

Although going to the football game is not a prerequisite for attending this event, we would like to get block seating for all of the Ph.D. students who are also planning to attend the Maryland game. If you would like to be a part of this group, please obtain your football coupon from the ticket office as soon as possible and give them to me by the RSVP deadline of Friday, October 17. My office is Room 310 in the old Management Building. If the door is locked, please write your name and email address on the back of your ticket and slide it under the door.

In addition to this, anyone who would like to help plan games and activities should email David Huang at dhuang@isye.gatech.edu or Jennifer Chung at gt8838b@prism.gatech.edu by Friday, October 17.

We are also seeking volunteers to help with food preparation. If you are willing to help, please contact Jim Luedtke at jluedtke@isye.gatech.edu.

We would like to maintain a strong sense of diversity as we continue the planning process, so please do not hesitate to lend a helping hand, regardless of background or experience.

Thank you very much, and I look forward to seeing you at the tailgate party!

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

However, the flow of shipments is often more complicated in practice. In an attempt to reduce sorting costs, load planners sometimes take this hub-and-spoke infrastructure and modify it considerably to maximize their truck utilization while satisfying service constraints. Decisions made by a load planner may have a cascading effect on load building throughout the network. As a result, decentralized load planning may result in expensive global solutions.

Academic as well as industrial researchers have adapted a hierarchical approach to design the hub-and-spoke networks: generate the hub-and-spoke network, route shipments within this hub-and-spoke network (generate a load plan) and finally, balance the empty trailers.

We present mathematical models and heuristics for each of the steps involved in the design of the hub-and-spoke network. The heuristics are implemented in a user-friendly graphical tool that can help understand patterns of freight-flow and provide insights into the design of the hub-and-spoke network. We also implemented the load planning sub-problem in a parallel computation environment to achieve significant speed-ups.

Because of the quick solution times, the tool lays the foundation to address pressing further research questions such as deciding location and number of hubs.

We have used data provided by Roadway Parcel Services, Inc., now FedEx Ground, as a case-study for the heuristics. Our solutions rival the existing industry solutions which have been a product of expensive commercial software and knowledge acquired by the network designers in the industry.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

on the issue of hedging such instruments and discuss how to bridge "traditional" credit valuation with "modern" credit valuation. Some practical issues around computing accurate hedge ratios are also discussed. ]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Dr. Huo will also discuss current happenings relating to ISyE-Statistics. Motivations and applications will be emphasized (he will try to make the talk less technical).

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

In order to do so, we first develop a general framework of convergence for optimization algorithms. Using this framework, we can show the convergence of traditional non-linear programming algorithms that have been suitably modified to use approximations of the objective function and its gradient. Then, we present one particular scheme for approximating the gradient and hessian of the objective function using linear regression. Finally, we describe a trust region algorithm that uses linear regression to form a linear or quadratic model of the objective function and provide computational results for such an algorithm run on the problems from the CUTE test set. ]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

a significant departure from current practice and the task of revamping RM technology to

incorporate choice models has been a challenging prospect. However, recent research advances

have now brought choice-based RM within striking distance of being practical. In this talk, we survey the recent research results in this area and discuss their implications for RM research and

practice.

ABOUT THE SPEAKER

Garrett Van Ryzin is the Paul M. Montrone Professor of Private Enterprise at the Columbia University Graduate School of Business. His research interests

include stochastic optimization, pricing and revenue management and supply chain management. He is co-author (with Kalyan Talluri) of the forthcoming book, The Theory and Practice of Revenue Management and is Editor in Chief of M&SOM. He received the B.S.E.E. degree from Columbia University, and

the degrees of S.M. (E.E.C.S.) and Ph.D in Operations Research from MIT.

Industrial and Systems Engineering

Contact Barbara Christopher

Among the questions we will seek to answer are, when is simulation an appropriate vehicle for analysis? Under what conditions should a new simulation be built? How does one design a model that addresses current research needs, while maintaining the flexibility for the future? What implementation technologies are available, and why choose a particular one? How does one validate a simulation whose purpose is to predict *future* results? How does one use a simulation to conduct an analysis? Along the way we will develop some insights into the aviation industry, including its probable future.

So join us for this talk

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

Dr. Jeffrey Tew will speak about the R&D efforts at the company.]]>

Industrial and Systems Engineering

Contact Barbara Christopher

The Pinwheel Problem, studied by several authors over the last 15 years, is a particular case of our problem when all job durations are equal. Our original interest in the Generalized Pinwheel Problem was motivated by radar sensor management applications. However, this is a natural scheduling problem that has other engineering applications. The studies of the Pinwheel Problem were primarily motivated by satellite communication applications.

We show that if the Generalized Pinwheel Problem is feasible, there exists a periodic schedule for this problem. We also provide necessary conditions for the feasibility, formulate an algorithm based on dynamic programming, and, since under certain conditions this problem is NP-hard, formulate and study heuristic algorithms. In particular, we model the Generalized Pinwheel Problem as a Semi-Markov Decision Process. This representation provides natural necessary conditions for the feasibility and leads to the development of heuristic algorithms that, according to simulations, outperform other heuristics.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

different quality of service (QoS) classes in a simple

telecommunications network and under an efficient network routing. The network uncertainties come from point-to-point demand processes that are driven by Brownian motions. Our model gives an analytical QoS pricing

formula in terms of perfect quality prices and the parameters of the point-to-point demand processes. In the absence of perfect quality prices, a utility based pricing approach is used. The model is illustrated with a numerical example and the QoS-price curve is found to

be S-shaped. ]]>

Industrial and Systems Engineering

Contact Barbara Christopher

Metso Automation specializes in automation and information management application networks and systems, field control solutions and life cycle services.

All ISyE grads and faculty are most welcome for the presentation, but due to IIE-Chapter's budgetary constraints, free food would be available only to IIE members.

]]>Industrial and Systems Engineering

Contact Barbara Christopher

Industrial and Systems Engineering

Contact Barbara Christopher

If you have interests in semiconductors and manufacturing, this is a talk you don't want to miss. A copy of the abstract follows.

]]>Industrial and Systems Engineering

Contact Barbara Christopher