2015 Informs Annual Meeting

MB59

INFORMS Philadelphia – 2015

MB58 58-Room 110A, CC Analytics in the Petrochemical and Petroleum Industries II Sponsor: ENRE – Natural Resources II – Petrochemicals and Petroleum Sponsored Session Chair: Tejinder Singh, Sr. Research Scientist, Delaware Research and Technology Center - Houston, TX, tejinder.singh@airliquide.com 1 - A Simulation and Optimization Framework for Petroleum Refinery Operations Ariel Uribe, Ecopetrol S.A., Km 7 Via Piedecuesta, Piedecuesta, Colombia, ariel.uribe@ecopetrol.com.co, Sandra Montagut, Omar Guerra In this work we present a framework for the simulation and optimization of petroleum refinery operations at strategic and tactic decision levels. Concerning the adequate modeling of processing units, the developed framework allows for the integration of economic models with both linear and non-linear empirical process models based on historical data, rigorous process simulators, or pilot plant data using scale up techniques. 2 - Long-term Demand Forecasting in Industrial Gas Markets Bin Yu, Air Liquide, 200 GBC Dr, Newark, DE, 19702, United States of America, bin.yu@airliquide.com, Adel Basli, Gildas Bonnier, Athanasios Kontopoulos, Brian Besancon In this talk we will focus on long-term demand forecasting of liquid oxygen and liquid nitrogen in the U.S. in 5-10 years. We analyzed the predictive performance of several forecasting techniques for IP in each sector using the employment and GDP as leading and associated variables. Moreover, we developed the method of decomposing the demand from the national level to local markets and identified the sectors that drive or restrain market growth in each local market. 3 - A Multi-period MINLP Model for Long-term, Quality-sensitive Shale Gas Development Markus G. Drouven, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, mdrouven@cmu.edu, Ignacio E. Grossmann In this work we address the long-term shale gas development problem which involves determining the optimal development strategy for drilling and fracturing gas wells, and designing a pipeline gathering infrastructure. The problem is formulated as a large-scale nonconvex MINLP involving concave investment costs and bilinear terms in the flow balances. We present a solution strategy that relies on an MILP approximation coupled with a restricted MINLP, which yields near optimal global solutions. MB59 59-Room 110B, CC Panel Discussion: The Impact of the Value-Based Approach on the Field of Strategy Cluster: Strategy Science Invited Session Chair: Nicolaj Siggelkow, University of Pennsylvania, 2000 SHDH, Philadelphia, PA, 19104, United States of America, siggelkow@wharton.upenn.edu 1 - The Impact of the Value-Based Approach on the Field of Strategy Moderator:Michael Ryall, University of Toronto, Rotman School, Toronto, ON, Canada, m.ryall.@mikeryall.com, Panelists: Peter Zemsky, Tomasz Obloj, Harborne Stuart Value-based strategy provides an intuitive, economic theory for business strategy. An immediate benefit is that it provides a coherent alternative to ad-hoc frameworks, and, in the process, it makes explicit the fact that profits are typically part of some larger economic pie. More broadly, by providing a theory for the economic aspect of strategy, it allows strategy research to focus on some of the richer issues in strategy, including, for example, organizational design, leadership, and execution. The discussion will also consider the empirical questions that arise from the general value capture model (i.e., bi-form games applied in the context of strategy). The mathematics indicate novel issues for empirical investigation. For example, whether the distinction between “competitive” vs “persuasive” resources is meaningful (as the model suggests it should be) and, if so, in which real-world settings one type is more efficacious than the other for superior returns.

2 - Regenerator Location Problem in Flexible Optical Networks Baris Yildiz, Bilkent University, Universiteler MAH., Ankara, 06800, Turkey, baris.yildiz@bilkent.edu.tr, Oya E. Karasan We present the regenerator location problem in flexible optical network that solves the regenerator placement, routing, bandwidth allocation and modulation selection problems jointly. We propose a novel branch and price algorithm for this challenging problem. Our results show that making routing, bandwidth allocation, modulation selection and regenerator placement decisions in a joint manner, it is possible to obtain drastic capacity enhancements with a limited regeneration capability. 3 - Risk Based Facility Location by using Fault Tree Analysis in Disaster Management Ibrahim Akgun, Assoc. Prof., Abdullah Göl University, Departmen of Industrial Engineering, Kayseri, 38080, Turkey, ibrahim.akgun@agu.edu.tr We develop an optimization model that minimizes the risk that a disaster-prone area may be exposed to because it is not supported by facilities located for prepositioning supplies. The risk is calculated as the multiplication of threat, vulnerability of the area, and consequence. The vulnerability is computed by using fault tree analysis and incorporated into the optimization model innovatively. The resulting non-linear integer program is linearized and solved as a linear integer program. 4 - Service System Design with Economies-of-scale and Congestion Samir Elhedhli, University of Waterloo, 200 University Avenue, Waterloo, Canada, elhedhli@uwaterloo.ca We formulate and provide solution methodologies for the service system design problem with immobile servers, stochastic demand, and capacity economies of scale. We start by reformulating the problem, and then provide solution approaches based on piecewise linearization, Second Order Cone Programming (SOCP), and Lagrangian Relaxation. Numerical results are provided MB57 57-Room 109B, CC Optimization of Power Systems Planning and Operation Sponsor: ENRE – Energy I – Electricity Sponsored Session Chair: Miguel Anjos, Polytechnique Montreal, Mathematics and Industrial Engineering, Montreal, Canada, miguel-f.anjos@polymtl.ca 1 - Chance-constrained Generation Expansion Planning Incorporating Bus Sensitivities William Rosehart, Schulich School of Engineering, University of Calgary, Calgary, AB, Canada, rosehart@ucalgary.ca, Monishaa Manick, Miguel Anjos A Generation Expansion Planning problem with load uncertainty is formulated based on joint chance-constrained programming. Sensitivities are used to allow greater emphasis to be placed on regions with high demand relative to generation, and similarly to allow for lesser emphasis on regions that are generation-rich. Numerical results are presented for IEEE test systems. 2 - Interrelationship Between Power Transmission and Storage Elements of the Power Network Enzo Sauma, Pontificia Universidad Catolica de Chile, Santiago, Chile, esauma@ing.puc.cl, Carlos Bustos, David Pozo, Javier Contreras, Sebastian De La Torre, Jose Aguado We study the interrelationship between the construction of new power transmission lines to integrate wind farms and the installation of new power- storage elements in the network. In particular, we analyze the effect of adding new power-storage components into the power system over the optimal transmission expansion plan. We illustrate our analysis using a stylized version of the Chilean main power system (Sistema Interconectado Central). 3 - Bilateral Contract Optimization in Power Markets Miguel Anjos, Polytechnique Montreal, Mathematics and Industrial Engineering, Montreal, Canada, miguel-f.anjos@polymtl.ca, François Gilbert, Patrice Marcotte, Gilles Savard We consider an energy broker linking its customers and the power grid through a two-sided portfolio of bilateral contracts. The contracts cover a number of actions taken by the customers on request within specified periods. Managing this portfolio raises a number of modelling and computational issues due to the aggregation of disparate resources. We propose an innovative algorithmic framework that models short-term decisions factoring in long-term information obtained from a separate model.

191