VENDOR WORKSHOPS

All workshops take place on Saturday, November 12, 2005, at the San Francisco Hilton.

You can register for any of these workshops when you register for the INFORMS Annual Meeting.

If you already registered for the meeting, click here for a faxable form to add these workshops.

AMPL Optimization
9:00am-5:00pm
Union Square 1

Linear, Nonlinear and Stochastic Programming using AMPL and Its Extensions
Cormac Lucas (CARISMA/Brunel University and OptiRisk Systems), Robert Fourer (Northwestern University and AMPL Optimization LLC) and Richard Waltz (Northwestern University and Ziena Optimization Inc.) have developed this full-day workshop on optimization modeling with AMPL, AMPL Studio and KNITRO. In the first part of the workshop, participants will gain an understanding of how to use the AMPL modeling language to formulate problems, to test algorithmic approaches and to implement decision support systems. The AMPL Studio user interface, modeling syntax and solver controls will be demonstrated, along with important aspects of linear and integer optimization models.

The next part of the workshop will provide a demonstration of introducing uncertainty into an optimization model and using risk constraints and risk decisions to enhance model formulations. The SAMPL stochastic extensions to AMPL and the SPInE stochastic modeling environment will be illustrated, using an asset and liability management problem as an example. The workshop will conclude with a description of the KNITRO solver, its several optimization methods and its key options. Convenient interfaces with modeling languages such as AMPL and with other software will be demonstrated. For a detailed list of the workshop sessions and topics, see http://www.ampl.com/MEETINGS/wkshpSched05.html.

Maximal Software, Inc.
9:00am-5:00pm
Continental Ballroom 2

Optimization Modeling with MPL: Advanced Modeling Techniques and New Trends for Stochastic Programming and Global Optimization
The workshop is divided into four parts:

• Optimization Modeling in MPL
• Connecting and Deploying Optimization Models
• Stochastic Programming
• Global Optimization

The workshop is designed to teach the participants how to develop their own optimization models and solve them using state-of-the-art, industrial-strength solvers. The workshop speakers come from both commercial and academic backgrounds, and have decades of experience in the field of optimization and modeling.

Bjarni Kristjansson, Maximal Software, Inc., will start in the morning with a quick introduction to the MPL Modeling System. This will be followed by a discussion of fundamental modeling techniques such as compound index sets, sparse data handling and how to formulate logical conditions with binary indicator variables. He will also demonstrate how to tune optimization solvers such as CPLEX, in addition to tricks and techniques for formulating models for better performance. Later in the morning, how to connect MPL to databases and spreadsheets will be covered, as well as how to embed optimization models into real-world applications.

Dr. Alan King, IBM Research, will discuss after lunch stochastic programming from the perspective of a practitioner who has developed an optimization model of an underlying resource allocation problem. He will present and discuss three approaches: robust optimization in the style of Bertsimas and Sim, stochastic programming with recourse and real options. Each of these approaches gives rise to a predictable reformulation of the optimization model, and so can be automatically generated by modeling language middleware.

Dr. Janos D. Pinter, PCS, Inc., will then present a concise introduction to global optimization (GO), including the most prominent model types and solution strategies. This will be followed by more details on the LGO solver suite and its MPL implementation. Finally, he will illustrate the usage of MPL/LGO by solving demo examples, and will comment on its application perspectives.

Participants will leave with an excellent overview of optimization modeling and a good working knowledge of how to formulate models for stochastic programming and global optimization. For more detailed information about the workshop please go to: http://www.maximalsoftware.com/seminar/SanFranNov05.

StatPoint, Inc.
9:00am-5:00pm
Union Square 5

STATGRAPHICS Centurion XV
With more than 150 statistical procedures, including ANOVA, regression, DOE, SPC, time series analysis, forecasting and much more. STATGRAPHICS Centurion has unrivaled capability to empower any operations research, R&D, quality improvement or Six Sigma program. The workshop will cover basic operation of STATGRAPHICS Centurion XV, including data management, performing analyses, use of StatFolios, modifying graphs, report generation and interfacing with other applications. The workshop will be an opportunity for users to experience the ease-of-use and power of STATGRAPHICS Centurion XV as well as provide an opportunity for potential clients to see first hand the unique time saving tools offered by STATGRAPHICS Centurion XV.

ILOG, Inc.
12:00pm-6:00pm
Continental Ballroom 1

ILOG Optimization Technologies Workshop
See new products and understand best approaches for your optimization needs. Join Irv Lustig and members of the ILOG CPLEX R&D team, for a special half-day technical workshop, followed by a social mixer / happy hour. Get an inside perspective and discuss problems you want to solve. This workshop will provide benefits for both beginners and advanced users alike – with new overview presentations, as well as advanced presentations on new technologies. The workshop content will cover:

The new ILOG OPL Development Studio 4
The fastest way to develop, tune and deploy CPLEX-based optimization models.

• Learn about the major advancements in the new ILOG OPL Development Studio 4

ILOG CPLEX
Delivering the fastest solve times achievable with the most reliable implementations of fundamental algorithms for solving linear programming, quadratic programming, quadratically constrained programming and mixed integer programming problems.

• See a special inside preview, from members of the R&D team, of the new upcoming release of ILOG CPLEX.

The workshop is free to all INFORMS participants. ILOG Optimization Technologies - simply incomparable. Come see why.

AIMMS – Paragon Decision Technology
1:00pm-5:00pm
Union Square 14

• The Pivot Table
• Extensions to the Generated Math Program library.
• The BARON solver upgrade and addition of the LGO solver
• Web Services

Go to http://www.aimms.com/aimms/downloads/aimms_36/new_features.html for more information about these new features.

At the end of the workshop, you should be able to get started with building your own application in AIMMS. The workshop forms the basis of what is required to create real-life applications in any business or industry. Materials will be available, and include a personal copy of AIMMS to get started right away. Registration is free for meeting attendees. To register, check the workshop box on the registration form or contact us directly at workshops@paragon.nl. Go to www.aimms.com for more information and a free trial copy of AIMMS.

Responsive Learning Technologies
12:00pm-5:00pm
Union Square 13

Online Games for Courses in Operations and Supply Chain Management
The following two structured sessions will each consist of an introduction and learning objectives, followed by a 45-minute game that participants will play in real time, and conclude with a debrief. Participants may leave early as necessary.

• Supply Chain Game: 1:30pm-2:45pm. Developed with faculty at Northwestern University for courses focusing on supply chain management.
• Littlefield Technologies: 3:30pm-4:45pm. Developed with faculty at Stanford University for introductory courses in operations management or production and inventory control.

Shorter software demonstrations and literature will be available any time during the workshop on a walk-in basis. For more details go to http://pom.responsive.net/informs.

Dash Optimization
1:00pm-5:00pm
Union Square 25

Fast Track Optimization Modeling using Xpress-MP
This workshop will provide participants with an insight into new modeling and optimization techniques, and first-hand accounts of leading business practitioners on how they are currently using optimization to solve practical problems. The workshop will feature modeling experts from Dash Optimization and Dash Optimization customers. For more information about this workshop please visit: http://dashoptimization.com/home/services/news/informs_sanfrancisco.html.

LINDO Systems, Inc.
1:00-5:00pm
Union Square 2

Exciting New Features in Modeling Systems from LINDO
We will describe and illustrate exciting new enhancements in the LINGO modeling language, the What’sBest spreadsheet optimizer and the LINDO API callable optimization library. Versions of LINDO API, as well as LINGO, are now available on a wide range of platforms in addition to Windows. Recent additions to the platform list are 64-bit versions for machines from AMD, Intel and Sun. Thus, large models are no longer constrained by the 2 to 3GB memory limitations of traditional Windows. A Macintosh version of the LINDO API is now available.

Notable new features in What’sBest are greatly expanded capability in classes of spreadsheet models that are well supported. Essentially, all mathematical functions of any kind in Excel are now supported. Continued development of the Global Solver, plus extensive industrial experience with it in What’sBest and the LINDO API, have brought it to the point that users may wish to make the Global Solver the default solver for general nonlinear programs. The Global Solver finds excellent solutions quickly, and if given sufficient time, proves optimality on arbitrary nonconvex models.

For LINGO, we will demonstrate (a) new programming language features in LINGO that are useful for coordinating and solving multiple models as part of one session, and (b) enhanced report generation features for easily generating formatted reports. A completely general debugging capability, applicable to integer programs as well as nonlinear programs, has been added to the LINDO API. For an infeasible model, it can quickly identify the typically half dozen constraints that constitute the infeasibility. The nonlinear global solver in the LINDO API has been extended so that it now analytically “understands” several dozen of the common mathematical functions, such as all trig functions, various probability distributions, common nonsmooth and discontinuous functions and various vector functions. It can also make improved use of information about analytical second derivatives.