Take advantage of these pre-conference workshops for a hands on demonstration of the latest in Analytics software. All attendees are welcome to join onsite or pre-register through the self-service center. Descriptions Below
Sunday April 14, 3-4:45pm, Room 304
Democratizing Supply Chain Prescriptive Analytics
The definition of supply chains has expanded over the years to include multiplier tiers of suppliers and customers. With more integrated business processes and capabilities to share and process information, competitive advantages lie in superior analytics – descriptive or predictive analytics are passé; prescriptive analytics and optimization are the differentiators to finding savings and locking in opportunities.
Integrated business processes can range from activities in procurement to final distribution. They include actions executed daily, such as allocations, shipments, scheduling, or inventory movements; or, decisions made strategically (weekly, monthly, quarterly, etc.), such as network design, multi-echelon inventory optimization or Sales and Operations Planning (S&OP).
AIMMS’ vision and mission of bringing the benefits of optimization to society has you covered regardless of your organization’s internal skills in, appetite for or maturity with optimization platforms.
Those with the appetite for Operations Research (OR) and skills in mathematical/algebraic modeling, can use AIMMS Developer to build bespoke models and apps. Those interested in deploying supply chain prescriptive analytic solutions directly into the hands of business users and decision makers, can use SC Navigator™ with its ready-to-configure-and-use apps (e.g. S&OP, Network Design, Center of Gravity, etc.).
In the first half of this workshop, we will demonstrate an S&OP application on the AIMMS SC Navigator™ platform. S&OP incorporates demand planning and supply/production planning, enabling management to make optimal decisions in an integrated manner. We will illustrate S&OP improvements compared with what most organizations use today (i.e. unrealistic unconstrained plans); we will also compare different optimization objectives, prioritization rules and business scenarios in the app.
In the second half, we will show how a similar app (as above) can be built from scratch – and customized infinitely – by organizations with OR skills using the AIMMS Developer platform.
Regardless of approach, we will show how the web enabled AIMMS PRO platform makes deployment extremely easy and intuitive – allowing the benefits of optimization to reach far corners of your operations!
Sunday, April 14, 9-10:45am, Room 303
Solve your first $500k problem in 75 minutes
Did you know that most analysts spend 80% of their time preparing and blending multiple data sources together and must rush through the analysis process to provide an end report? If this sounds like a familiar exercise, join this session so you can break free from the drudgery of prep and blend to turn hour long tasks into minutes. We will dive into data for a true analysis to provide insights to alter organizations and show you to thrive at analytics instead of just surviving. During this session you will build your first repeatable workflow to solve that half a million dollar problem before lunch!
Sunday, April 14, 3-4:45pm, Room 303
Adding Optimization to Your Applications, Quickly and Reliably: From Prototyping to Integration with AMPL
Optimization is the most widely adopted technology of Prescriptive Analytics, but also the most challenging to implement:
* How can you prototype an optimization application fast enough to get results before the problem owner loses interest?
* How can you develop optimization-based procedures to get results you can use, within your time and resource requirements?
* How can you integrate optimization into your enterprise’s decision-making systems?
In this presentation, we show how AMPL gets you going without elaborate training, extra programmers, or premature commitments. We start by introducing model-based optimization, the key approach to streamlining the optimization modeling cycle and building successful applications today. Then we demonstrate how AMPL’s design of a language and system for model-based optimization is able to offer exceptional power of expression while maintaining ease of use.
The remainder of the presentation takes a single example through successive stages of the optimization modeling lifecycle:
* Prototyping in an interactive command environment.
* Development of optimization procedures via AMPL’s built-in scripting language.
* Integration through APIs to widely used programming languages, including C++, C#, Java, and MATLAB, and featuring the popular data science languages Python and R.
Our example is simple enough for participants to follow its development through the course of this short workshop, yet rich enough to serve as a foundation for appreciating model-based optimization in practice.
Sunday, April 14, 1-2:45pm, Room 303
Applying AnyLogic and AnyLogic Cloud to Solve Various Business Challenges
Presented by: Dr. Andrei Borshchev, CEO & Nikolay Churkov, Head of Software Development
After a quick tour around the AnyLogic model development environment, we will have a live model building session. We will build a manufacturing model and showcase the new features of AnyLogic Material Handling Library such as AGVs moving in free space. We will then upload the model to AnyLogic Cloud and set up and run various experiments with the web UI. We will explain the benefits of having simulation models/digital twins running in cloud-based execution environment. At the end, as always, we will have time for Q&A.
Sunday April 14, 3-4:45pm, Room 305
Nonlinear optimization in a nutshell with Artelys Knitro
Nonlinear optimization is used in many applications in a broad range of industries such as economy, finance, energy, health, 3D modeling, and marketing. With four algorithms and great configuration capabilities, Artelys Knitro is the leading solver for nonlinear optimization and demonstrates high performance for large scale problems. This session will introduce you to Artelys Knitro, its key features and modeling capabilities, with a particular emphasis on the latest major improvements including a new algorithm designed for second order cone constraints, and a new API to build-up models piece-by-piece with structural information. Finally, this presentation will end with an insightful session on nonlinear modeling tips and tricks!
Sunday, April 14, 11am-12:45pm, Room 304
How To Get Started with Automated Machine Learning
Presented by: Jen Underwood
Today there is way too much data to manually analyze and a data scientist shortage. Stop waiting for data scientists and learn how you can accelerate insight to action with no-code DataRobot automated machine learning with your existing analytics talent.
In this session, we will introduce automated machine learning and teach you how to get started using DataRobot. We will cover how to select appropriate business problems to solve, how to prepare data, analyze data, build a machine learning model and put it to work for the business in dashboards or applications. Lastly, we will provide tips for effectively translating quantitative insights and telling a compelling story throughout the entire project life-cycle.
Workshop participants will:
– Walk-through how to structure machine learning projects
– Get an introduction to DataRobot automated machine learning
– Learn how to effectively communicate results to the business
Speaker Bio: Jen Underwood is a Senior Director at DataRobot. She has
a unique blend of product management and “hands-on” experience in data warehousing, reporting, visualization, and advanced analytics. In addition to keeping a constant pulse on industry trends, she enjoys digging into oceans of data to solve complex problems with machine learning.
Over the past 20 years, Jen has held worldwide product management roles at Microsoft and served as a technical lead for system implementation firms. She has experience launching new products and turning around failed projects. Most recently she provided advisory, strategy, educational content development, and marketing services to 100+ technology vendors through her own firm. She has been mentioned by KD Nuggets, Information Management and Forbes for her work. She also has written for InformationWeek, O’Reilly Media, and numerous other tech industry publications.
Jen has a Bachelor of Business Administration – Marketing, Cum Laude from the University of Wisconsin, Milwaukee and a post-graduate certificate in Computer Science – Data Mining from the University of California, San Diego. She was also honored to be a former IBM Analytics Insider, Tableau Zen Master, and Top 10 Women Influencer.
Sunday, April 14, 11am-12:45pm, Room 306
Map-Reduce for Optimization: Build and Deploy Optimization Decomposition on Distributed Resources in the Cloud
Presented by: Filippo Focacci, Co-founder, CEO
DecisionBrain Optimization Server (DBOS) seamlessly runs multiple CPU intensive computational jobs locally or remotely and provides administrative tools to easily monitor and manage them.
Specifically designed to help build and deploy fully scalable optimization applications, it enables optimization developers to easily design parallel and distributed decomposition methods and large neighborhood search methods.
In this workshop, we demonstrate how this technology can be used to solve a large-scale optimization problem by mimicking, in a very simple way, a Map-Reduce type of approach: split the problem into sub-problems, solve each sub-problem on a different distributed resource, and eventually, combine the different results.
Before founding DecisionBrain, Filippo worked for ILOG and IBM for 20 years where he held several leadership positions in Consulting, R&D, Product Management and Product Marketing in the areas of Supply Chain, Logistics and Optimization. He received a Ph.D. in Operations Research (OR) from the University of Modena (Italy) and has over 20 years of experience applying OR techniques in industrial applications in several optimization domains. He has published several Supply Chain and Optimization articles for international conferences and journals. He has been granted a patent for Optimization Models.
Sunday, April 14, 11am-12:45pm, Room 305
Decision Framing Made Easy
Presented by: Jeremy Walker, V.P.
Interested in Decision Framing? Come participate in the creation and selection process of a compelling strategy surrounding the shipping of bio-diesel by a pipeline company, presented as a case example. As a group, you will be exposed to the soft skills of facilitation and the step-by-step sequence of creating a Decision Frame, collecting and categorizing issues, building a Decision Hierarchy, designing Strategy Tables, weaving together a set of different thematic choices and ultimately, comparing your Strategies. You will see the latest version of DTrio, a Decision Framing tool, generate the Frame and TreeTop, a Decision Analysis tool, run uncertainty analysis using Tornados, Decision Tree and Cumulative Probability curves.
Sunday, April 14, 3-4:45pm, Room 306
Calling all OR Heroes: Learn How to Model and Deploy Solutions in 75% Less Time with FICO® Xpress Insight
Presented by: Horia Tipi
Join our hands-on workshop to experience how FICO® Xpress Insight helps you read data in any format from any source, integrates with your own machine learning and solvers (or Xpress Solver), enables collaboration with business users, deploys decision support or automated solutions—and does it all in 75% less time.
- Get the overview of the great new features of FICO® Xpress Solver, Mosel, Workbench, and Insight.
- Discover the new View Editor, which reduces GUI development times from minutes to seconds.
- Experience Xpress Insight first-hand, including the power of Xpress Mosel—the premier mathematical modeling, analytic orchestration and programming language.
- Learn through real-world business examples how developers integrate their own solver(s) with the powerful Xpress Insight platform.
- See how the flexible, free-to-use Xpress Mosel modeling and programming language allows you to make your solver available to thousands of other Mosel users.
All attendees will receive a link to the complimentary FICO® Xpress Optimization Community License.
Sunday, April 14, 9-10:45am, Room 305
Creating Interactive Analytics on the Web with Forio Epicenter
Presented by: Michael Bean
Forio Epicenter makes your model available to hundreds of people within your organization through the browser. Forio Epicenter supports Excel, R, Python, Julia and other languages for optimization, machine learning, simulation, and other analytic techniques. The platform is enterprise-compatible with the ability to integrate with an organization’s existing IT infrastructure and tiered control for thousands of users. During this workshop, we will introduce Epicenter and share sample, interactive online models. Then we will show you how to create an publish an online interactive analytics application.
Sunday, April 14, 3-4:45pm, Room 301
Get Results from Analytic Models in Excel, Dashboards, and Business Systems
Presenter: Daniel Fylstra, President, Frontline Systems Inc.
See the easiest way to build models, get analytics results, and share them across your organization at this workshop, where business people, developers, and experts are all welcome. We’ll demonstrate how you can learn analytic modeling, leveraging skills you already have, as you work to quickly get insights and bottom-line results from data and text mining, Monte Carlo simulation and risk analysis, conventional and stochastic optimization models.
We’ll explain how you can leverage Office 365 desktop and cloud, and how to get the data your models need, whether it’s in spreadsheets, databases, BI systems, cloud sources, or big data clusters. We’ll show how to easily deploy and run your model in dashboards like Tableau and Power BI, in live business processes with Microsoft Flow, and in your own server, web or mobile applications. You’ll see why over 9,300 organizations have used Frontline Solvers over more than 25 years, and why over 530,000 users are already using our cloud analytics apps.
Sunday, April 14, 9-10:45am, Room 302
Deploying GAMS models with GAMS MIRO
Presented by: Franz Nelissen, Hermann von Westerholt
Are you a modeling expert and would like to run your GAMS program from within a web browser, being able to visualize input and output data? Or an enterprise that requires a graphical user interface so that their planners can work more productively, leveraging the full potential optimization brings?
During this tutorial, we will introduce GAMS MIRO, a web interface for your GAMS models. We will start explaining the fundamental concepts of GAMS and the advantages of using GAMS to build optimization-based decision support applications. We will continue with the various options to connect GAMS models to other application and briefly cover some recent developments.
The central part of the tutorial will be about GAMS MIRO. It has tight connections to the GAMS modeling system that allow you to perform data manipulation, scenario management, graphical evaluation of the results and much more from within the web browser by adding very few annotations to your model. You specify which input and output datasets you want to visualize, and the result is a fully functional GUI that can be launched directly from the new GAMS Studio or via a shortcut on your Desktop. GAMS MIRO also facilitates the generation, organization, and sensitivity analysis of multiple scenarios of an optimization model. A server version will support features as managing multiple optimization models for concurrent users with access management, load balancing, batch configuration and much more.
Sunday, April 14, 1-2:45pm, Room 304
Recent Developments in the Gurobi Optimizer
Presented by: Ed Rothberg (CEO), Dan Jeffrey (Sr. Support Engineer), Gwyneth Butera (Sr. Support Engineer)
Learn about the latest advances in Gurobi Optimizer. In this workshop, we’ll give an overview of our latest Gurobi 8.1 release, which includes performance improvements, new features, and other enhancements. We’ll also give live demos of our Compute Server and Instant Cloud products, specifically showing how to seamlessly transition workload between them in order to better manage peak demand or cope with failures. Finally, we’ll demonstrate our new Optimization Application Kit, which is designed to help illustrate and demystify the process of building scalable optimization applications.
Sunday, April 14, 11am-12:45pm, Room 301
Recent Advances on Large Scheduling Problems in CP Optimizer, Multiobjective Optimization in CPLEX
Presented by: Xavier Nodet – Senior Software Engineer Manager, CPLEX Optimization Studio, IBM Hybrid Cloud, Ferenc Katai – Offering Manager (OM) – CPLEX/CPO/OPL = IBM ILOG CPLEX Optimization Studio (former OPL Studio), IBM Hybrid Cloud & Virginie Grandhaye – Decision Optimization Offering Manager, IBM Hybrid Cloud
After recapping the main principles of CP Optimizer, this talk will focus on recent advances in the automatic search allowing to handle some scheduling problems with hundreds of thousands of activities in a matter of seconds.
The upcoming CPLEX release features multi-objective optimization. Available for LPs and MIPs, it allows to specify combinations of hierarchical and blended objectives, and gives you an optimal solution for your instance. Thanks to tolerances on each sub-objective, you can evaluate the impact that each objective has on the others. And the weights on each objective allow to scale each objective, either for trade-off exploration, or to help numerical stability.
What’s new about Decision optimization for Watson Studio, our unified environment that allows you to easily create and deploy Decision Optimization and Machine Learning models to address your business challenges. We’ll give you an update on roadmap items.
Sunday, April 14, 3-4:45pm, Room 302
Yes, You Can Add Optimization to Your Large Complex What-If Excel Models
Presented by: Linus Schrage
Ease-of-use and a powerful library of functions have helped make Excel the most widely used tool for building planning models. Learn how to get globally optimal solutions to your planning models using convenient Excel functions such as VLOOKUP, SUMIF, MAX, ABS, MATCH, INDEX; How to handle IFERROR, and more. Using examples from refineries, supply chains, sourcing, routing, and more, we will show you how to turn your large, complex What-If spreadsheet models into optimization models, using What’sBest! and LINGO from LINDO Systems.
We will cover the various model types such as blending, financial portfolios, multi-period planning, cutting stock, etc. and suggest approaches that have worked well for our customers over the years. For modeling financial portfolios, there are special functions for computing the variance of a portfolio, as well as for estimating the covariance matrix from raw data. For modeling oil refineries, there are special functions for modeling heat exchangers and , for modeling the behavior of gases, e.g., “steam tables”, and modeling the nonlinear behavior of blends of hydrocarbons. For modeling combinatorial problems, support for cardinality constraints, and AllDiff functions, popular in constraint programming, is provided.
Sunday, April 14, 11am-12:45pm, Room 302
Data Analytics with MATLAB
MATLAB makes it easy to do descriptive, predictive and prescriptive analytics with tools to access and preprocess data, build machine learning and optimization models, and deploy models to enterprise IT systems. In this workshop, we’ll demonstrate these capabilities through explorations of deep learning and optimization.
Deep learning is a type of machine learning that can achieve state-of-the-art accuracy in many human-like tasks such as naming objects in a scene or analyzing sentiment in text. We’ll use MATLAB features that simplify creating and training deep neural networks without a need for low-level programming. In doing so, you will gain practical knowledge of the domain of deep learning.
Building an optimization model is just one part of a successful prescriptive analytics project. We will show you how you can build and deploy a web app with MATLAB for selecting sites for disaster relief. We will use MATLAB to process Big Data, visualize it with charts and maps, build a predictive model, and use those results as input to the optimization model.
Sunday, April 14, 11am-12:45pm, Room 303
Turning Algorithms into Applications to put AI in the Hands of Enterprise Teams Faster than Ever
Presented by: Ronan O’ Donovan, Partner, Opex Analytics
AI in Action
The hardest part of integrating AI in organizations is translating strategic plans into practical applications. In this workshop, we will walk you through Opex Analytics’ best practices for converting algorithms into applications. We will showcase our platform, but also walk you through other common approaches – both commercial and open sources. What works best for you will depend on where you are in your own AI journey.
Don’t miss this opportunity to learn more about the business value being created on the front lines of enterprises through best in class deployment of AI.
The Opex Platform
Rapid AI implementation across the enterprise
Our Platform abstracts away the complexity of deploying AI solutions by delivering modern web applications that harness our custom algorithms and data models.
Speed and Agility
Opex Analytics brings a fundamentally different approach, ensuring enterprises experience the business benefits of AI earlier and faster. After we create unique algorithms and models to solve your problem, your business gets a fully-designed product in minutes, drastically reducing the time and cost to deploy, iterate and improve solutions.
Easy to Use
With the Opex Analytics Platform, we turn AI solutions into modern, usable web applications. Enable your teams to work faster and better through products that are simple, intuitive, robust and tailored to meet specific business needs.
Sunday, April 14, 9-10:45am, Room 301
Methodologies for Deploying applications that combine machine learning tools and optimization solvers (CPLEX/ ODH|CPLEX)
Presented by: Robert Ashford & Alkis Vazacopoulos
Organizations are increasingly hiring Data Scientists with Open Source skills. They leverage the capabilities and work with Open Source tools like R, Python, Spark, as well as integration with CPLEX and large data. Come learn how to integrate with Open Source tools to enable clients to get the best of both worlds (Open Source programming and the Modeler GUIs for those who prefer not to code).
Furthermore we will review the latest developments/results in CPLEX Optimization Studio and the new ODh+CPLEX.
- Combining Machine Learning & Prescriptive Analytics with CPLEX 101 – Alkis Vazacopoulos, Optimization Direct & Sumeet Parashar, IBM
- Build and deploy scalable CPLEX applications with DecisionBrain DBOS — quickly, easily, and efficiently. – Filippo Focacci, DecisionBrain
- Solving Hard Mixed Integer programming problems with ODH|CPLEX Robert Ashford, Optimization Direct
- Leveraging python packages to improve model-building performance – Joshua Woodruff, Optimized Financial Systems
Sunday, April 14, 1-2:45pm, Room 305
You Can Sell and Deploy Successful Optimization Solutions
Presented by: Irv Lustig, PhD, Optimization Principal
Most optimization practitioners have a good understanding of how to build a model, and write code to integrate that model with data to solve an optimization problem. However, challenges remain for many practitioners when it comes to having those models used in practice. The challenges are first with respect to selling the project, and then secondly, deploying the model into a real application. Based on years of delivering optimization projects and using a number of real-world examples, Dr. Lustig will describe a number of best practices used by Princeton Consultants to sell and deploy optimization solutions. This workshop is appropriate for both technical and non-technical audiences.
Dr. Irv Lustig is the Optimization Principal at Princeton Consultants, where he leads development of custom optimization software solutions. Previously, Irv was an IBM/ILOG/CPLEX employee from 1993-2014, where he variously led or managed technical services, sales and product development, and researched forecasting and optimization integration. Irv received Sc.B. and Sc.M. degrees in Applied Mathematics/Computer Science from Brown University and a Ph.D. in Operations Research from Stanford University. He has authored more than 30 articles and scientific papers and has received the Beale-Orchard Hayes Prize for excellence in computational mathematical programming. Irv is a Certified Analytics Professional as recognized by INFORMS.
Sunday, April 14, 1-2:45pm, Room 301
Solving Business Problems with SAS Analytics and OPTMODEL
Presented by: Rob Pratt, Senior R&D Manager Pelin Çay, Senior Associate OR Specialist Ed Hughes, Principal Product Manager
SAS offers diverse analytic capabilities, including data integration, statistical analysis, data and text mining, machine learning, artificial intelligence, forecasting, optimization, scheduling, and simulation. The OPTMODEL procedure from SAS provides you with a full-featured optimization modeling language, access to a diverse set of solvers, and the ability to create and use customized solution algorithms. SAS analytic capabilities are also available through the cloud-enabled, open design of SAS® Viya®. You can program in SAS or in other languages—Python, Lua, Java, and R.
We’ll explore analytical and optimization case studies drawn directly from our work with SAS users in a wide range of industries. These case studies demonstrate PROC OPTMODEL’s power and versatility in building and solving optimization models and illustrate how deeply optimization integrates with the full array of analytics provided by SAS.
Sunday, April 14, 9-10:45am, Room 304
Manipulating Data and Analytics Using SAS® University Edition
Presented by: Tom Grant
This introduction to SAS University Edition hands-on workshop shows how one can use the menu driven tasks and SAS code in SAS University Edition 4.3 to perform common reporting and research tasks: querying, reporting, and analyzing data. Several statistical procedures will be used to analyze data and produce reports. SAS University Edition provides a SAS graphical point-and-click interface as well as code that helps you exploit the power of SAS and publish dynamic results in a Microsoft Windows client application. Demonstrations in the presentation will use research type data and tasks in illustrating the functionality of SAS University Edition.
Learn how to In this course you will learn to access your data, combine tables, compute new variables, explore data with simple statistics and graphs, and perform sophisticated statistical analyses with SAS University Edition. This course does not teach statistical concepts, but teaches how to use these tools in SAS University Edition.
Tom Grant holds a Master of Science degree in Operations Research from Virginia Commonwealth University. Prior to joining SAS in 2000, Tom worked in the accounting department of national furniture retailer, and developed statistical models to predict inventory loses. Tom was also manager of the Marketing Analytics Group for a retail clothing catalog company, in charge of building predictive response and life-time value models. Since joining SAS, Tom has worked as a consultant, assisting SAS customers with implementation of analytic projects in various industries, including banking, retail services, insurance, manufacturing, and web-site analysis. Tom also teaches Intro to Statistics at a small liberal arts college. Tom is currently a principal analytical training consultant in the SAS Global Academic Program and he assists higher education institutions in the effective use of SAS.
Sunday, April 14, 1-2:45pm, Room 302
JMP and the Predictive Modeling Workflow
Presented by: Kevin Potcner
The amount of data that the modern data analyst has access to continues to grow. This provides a great opportunity for analysts to utilize some of the more modern predictive modeling techniques. A typical real-world predictive modeling workflow includes data cleaning and exploration, model fitting, model validation, model comparison, final model selection and deployment of the final predictive model.
In this session we illustrate the predictive modeling workflow by analyzing a real dataset. After data preparation and initial exploration, we will create a number of predictive models such as multiple linear regression, regression tree, Neural Net, and K-Nearest Neighbors.
We will “publish” each of the models to the Formula Depot, and explore and select the best model using the Prediction Profiler and JMP’s Model Comparison tool.
Code will be created in a variety of programming languages (e.g., SAS, SQL, Python, et al.) in order to implement that model in a production environment.
Sunday, April 14, 1-2:45pm, Room 306
New Innovations: Cloud Computing, Real-Time Scheduling, Industry 4.0, and More
Presented by: Renee Thiesing and Katie Prochaska (Simio LLC)
With Simio leveraging the cloud computing power of Microsoft Azure to support your most demanding applications, its compatibility with Schneider Electric’s Wonderware to allow detailed production scheduling with real-time data and risk analysis, and ability to schedule and reschedule in real-time, Simio is leading the way in Industry 4.0 and creating a digital factory. Outside our immense technology partner advances, we have great new features, application areas and capabilities! Come explore an overview of the new Simio experience and see why we are always “Forward Thinking.”