A datadriven robust optimization approach to scenariobased. A data driven robust optimization approach to operational optimization of industrial steam systems under uncertainty. Datadriven robust optimization for steam systems in. Datadriven robust optimization based on kernel learning. Theory and applications of robust optimization dimitris bertsimas. Distributionally robust optimization dro online linear programming olp least squares with nonconvex regularization lsnr alternating direction method of multipliers admm data. Furthermore, the book has been structured so that the first part could easily be used as the text for a graduate level course in robust optimization. A novel generative adversarial network gan based data driven distributionally robust chance constrained. Professor shahab mohaghegh, being one of the most innovative and experienced thought leaders in the field of datadriven modeling in the upstream, has written a comprehensive and readable book that finally puts to bed the persistent complaints in the industry. The primary goal of this preface is to provide the reader with a. I thought one cant exist without the other and they must imply the same. The application of robust optimization in power systems. Datadriven robust optimization mathematical programming.
This electronic textbook is a studentcontributed opensource text covering a variety of topics on process. Momentbased uncertainty sets momentbased uncertainty sets is comparable to that of the classical. May 15, 2017 watch this webinar and understand the basics of robust optimization, and why there is a difference between an optimal setpoint and a robust setpoint. Outline 1 general overview 2 static problems 3 adjustable ro 4 twostages problems with real recourse 5 multistage problems with real recourse 6 multistage with integer recourse michael. While there are many highoptimization ro came from the robust control community, ro is a distinct. The last decade witnessed an explosion in the availability of data for operations research applications. Data driven robust optimization moahammad namakshenas and mir saman pishvaee, school of industrial engineering, iran university of science and technology, tehran, iran chapter 2. Second, the existing studies focus on either data driven static robust optimization or data driven aro, without deriving a data driven uncertainty set that is general enough for both robust optimization paradigms.
Recently, probabilistically robust optimization has gained popularity by the introduction of rigorous theories such as scenario optimization able to quantify the robustness level of solutions obtained by randomization. The growing data driven optimization approaches offer new techniques to eliminate uncertainty in the process system engineering community. Datadriven robust optimization with application to. Datadriven robust optimization based on kernel learning chao shanga, xiaolin huangb, fengqi you,a asmith school of chemical and biomolecular engineering, cornell university, ithaca, new. If you use this work for academic purposes, please cite barath, d. A datadriven robust optimization approach to scenariobased stochastic model. Datadriven distributionally robust optimal power flow for. Datadriven reservoir modeling introduces new technology and protocols intelligent systems that teach the reader how to apply data analytics to solve realworld, reservoir engineering problems. Distributionally robust optimization reference delage, e. Examples include the case of airlines selling ight tickets, hotels trying to book.
Introduction in many industries, managers are faced with the challenge of selling a xed amount of inventory within a speci c time horizon. Distributionally robust optimization under moment uncertainty with application to datadriven problems. Recently, probabilistically robust optimization has gained popularity by the introduction of rigorous theories such as scenario optimization able to quantify the robustness level of. Focusing on one of the centers infusion units roughly 90 daily. Cooperative datadriven distributionally robust optimization. The approach is flexible and widely applicable, and robust optimization problems built from our new sets are. In this paper, we propose an e ective data driven approach for robust optimization. A datadriven robust optimization approach to operational optimization of industrial steam systems under uncertainty. This book presents a comprehensive theoretical treatment of the h2 approach to datadriven control design. Datadriven methodologies have recently emerged as an important paradigm alternative to modelbased controller design and several such methodologies are formulated as an h2 performance optimization. Data driven robust optimization gtest are based on phidivergences see also 32. Michael poss introduction to robust optimization may 30, 2017 9 53.
The models proposed in this paper fall within the scope of datadriven distributionally robust optimization 7, 16, 20, 23. Data driven optimization models and algorithms youtube. This paper presents a novel deep learning based datadriven optimization method. The manner in which the joint problem has been handled in recent studies has varied. This book is devoted to robust optimization a speci. This may somewhat explain the scarce of applications of machine learning models in robust optimization all this time. Welcome to the northwestern university process optimization open textbook. Datadriven distributionally robust optimization using the 119 amongallmetricbasedambiguitysetsstudiedtodate,thekullbackleiblerambiguity set has attracted.
The application of robust optimization in power systems final project report. One popular approach to addressing the distributional uncertainty, known as the distributionally robust optimization, is to hedge against the worst case over an uncertainty set of candidate. There are several ways to apply robust optimization and the choice of form is. Our analysis shows that the lro approach is fully data driven, and it takes advantage of the full strength of the available data while maintaining a certain level of robustness. Datadriven robust optimization for steam systems in ethylene. Datadriven robust optimization gtest are based on phidivergences see also 32. Likelihood robust optimization for datadriven problems. A compact convex uncertainty set can be efficiently learnt from massive data.
Robust optimization is still a relatively new approach to optimization problems affected by uncertainty, but it has already proved so useful in real applications that it is difficult. This entails the development of a datadriven robust optimization approach with noveluncertaintysets. Data driven distributionally robust optimization using the wasserstein metric. Computational tractability is well preserved by the robust counterpart problem.
Dec 31, 20 the last decade witnessed an explosion in the availability of data for operations research applications. Data driven robust optimization moahammad namakshenas and mir saman pishvaee, school of industrial engineering, iran university of science and technology, tehran. The models proposed in this paper fall within the scope of data driven distributionally robust optimization 7, 16, 20, 23. The purpose of this tutorial is to present a mathematical framework that is wellsuited to the limited information available in reallife problems and captures the decisionmakers attitude. A datadriven robust optimization ddro methodology is proposed to deal with uncertainty in the optimization of steam system in an ethylene plant.
In robust optimization, random variables are modeled as uncertain parameters belonging to a convex uncertainty set, and the decision maker protects the system against the worst case. The probability distribution of the random variable. A datadriven approach to choosing machine learning algorithms. Pishvaee,data driven robust optimization, in robust and constrained optimization. They focus on the case where the uncertain parameters u. Data driven distributionally robust optimization using the wasserstein metric 3 computational e ort is reported to be considerable, and there is no guarantee to nd the global optimum. Datadriven robust optimization with application to portfolio management, working papers 2014104, department of research, ipag business school. Nevertheless, tractability results are available for special cases. However, after reading on internet and your blog i understand that machine learning is the study and development of algorithms that can learn from data and make predictions. A data driven robust optimization ddro methodology is proposed to deal with uncertainty in the optimization of steam system in an ethylene plant. Ddr leverages the tractability of robust optimization and enables one to explicitly. Distributionally robust optimization dro online linear programming olp least squares with nonconvex regularization lsnr alternating direction method of multipliers admm data drivenoptimization yinyuye k.
Fraction of data coverage can be conveniently adjusted using only one parameter. Datadriven decision making under uncertainty integrating. Data driven reservoir modeling introduces new technology and protocols intelligent systems that teach the reader how to apply data analytics to solve realworld. Furthermore, the book has been structured so that the first part could. Our analysis shows that the lro approach is fully datadriven, and it takes advantage of the full strength of the available data while maintaining a certain level of. Data driven optimization uses observations of the random variables as direct inputs to the mathematical programming problems. Datadriven reservoir modeling ebooks spe book store.
Brown y, constantine caramanis z july 6, 2007 abstract in this paper we survey the primary research, both theoretical and applied, in the. A datadriven robust optimization framework was developed for automatically incorporating correlation and distributional asymmetry information from uncertainty data into robust. Efficient recovery of tangent planes in stereo images. The argument is based on the fact that a separate analysis allows obtaining. These methods are also relevant to data driven optimization methods. Instead of immunizing optimal solutions against worstcase observations that are available data robust methods distributionally robust optimization takes the worstcase over a family. This project has focused on the application of robust optimization for power system operations and operational. Datadriven distributionally robust optimization using the wasserstein metric 3 computational e ort is reported to be considerable, and there is no guarantee to nd the global. The growing datadriven optimization approaches offer new techniques to eliminate uncertainty in the process system engineering community.
Robust optimization methodology and applications 455 cones k1. Motivated by this growing availability, we propose a novel schema for. This limitation hinders them from having a wideranging application scope. Data driven robust optimization 7 delage and ye 8 prov ed that the robust counterpart of the problem 2 according to u can be formulated as a semidefinite program sdp.
Theory and applications of robust optimization siam. In robust optimization, random variables are modeled as uncertain parameters belonging to a convex uncertainty set, and the decision maker protects the system against the worst case within that set. Distributionally robust optimization under moment uncertainty with application to data driven problems. Chapter 4 robust field development optimization using the. The purpose of this tutorial is to present a mathematical framework that is wellsuited to the limited information available in reallife problems and captures the decisionmakers attitude towards uncertainty.
Closest in spirit to our work is the robust sample average approximation, which seeks decisions that are robust with respect to the ambiguity set of all distributions that pass a prescribed statistical hypothesis test. Datadriven robust optimization with application to portfolio. Robust optimization problems where theuncertainty isaprobability. Motivated by this growing availability, we propose a novel schema for utilizing data to design uncertainty sets for robust optimization using statistical hypothesis. Watch this webinar and understand the basics of robust optimization, and why there is a difference between an optimal setpoint and a robust setpoint. Cooperative datadriven distributionally robust optimization ashish cherukuri jorge cort. Datadriven robust optimization massachusetts institute of. Datadriven robust optimization with application to portfolio management, working papers 2014104. Totestforpracticalperformance,weleverageauniquedatasetfroma cancer center that combines realtime locations, electronic health records, and appointments log. Keywords robust optimization datadriven optimization chanceconstraints.
Data driven optimization models and algorithms simons institute. Datadriven appointmentscheduling under uncertainty. In this paper we develop a method of data driven stochastic programming that avoids the arti. This book is an authoritative reference that will be very useful to researchers working in this area. This paper presents a novel deep learning based data driven optimization method. Brown y, constantine caramanis z july 6, 2007 abstract in this paper we survey the primary research. The estimation method is typically selected with the goal to achieve maximum prediction accuracy but without tailoring it to the optimization problem at hand. Datadriven distributionally robust optimization using the wasserstein metric. A novel generative adversarial network gan based datadriven distributionally robust chance. Datadriven distributionally robust optimization using the. Robust optimization is a distinct approach to optimizations problems that allows for the incorporation of uncertainty.
709 244 893 287 1381 257 689 1487 1392 136 415 419 551 775 153 1188 214 1195 28 568 241 71 1148 1037 482 449 503 391 1414 584 1213 1163 360 694 194 482 746 431