Informs Annual Meeting 2017

MC25

INFORMS Houston – 2017

MC23

3 - Multi Criteria Decision Analysis for Contingency Base Siting Jeffrey Cegan, USACE, 676 Virgiia Road, Concord, MA, 01742, United States, jeffrey.c.cegan@usace.army.edu, Matthew Bates The US Army is in need of the ability to rapidly plan and site base camps in order to maximize the combat effectiveness of deployed forces. By defining mission specific requirements prior to the deployment of base camps, commanders are able to optimize base camp site selection. Currently, existing doctrine and guidance for contingency basing (CB) are not developed to support deployment in dense urban environments. Preliminary research reveals the inadequacy of current doctrine for conducting operations in such environments (Chief of Staff of the Army, 2013). The CB-SITE project seeks to address these problems by utilizing multi-criteria decision analysis for base camp site suitability assessment. 4 - Any Cognitive Bias in Spatial Decision Analysis Processes? Valentina Ferretti, London School of Economics and Political Science, Houghton Street| London | WC2A 2AE, London, United Kingdom, V.Ferretti@lse.ac.uk The need and interest to consider cognitive and motivational biases has been recognized in different disciplines (e.g. economics, decision theory, finance, risk analysis, to name the most relevant ones) and has recently reached environmental decision making. Within this domain, the intrinsic presence of a spatial dimension of both alternatives and criteria calls for the use of maps throughout decision analysis processes. This talk presents insights from a literature review on cognitive biases in spatial decisions, as well as some preliminary results from a behavioural experiment on the spatial dimension of biases. 350A Control and Optimization Techniques for Power Systems II Invited: Energy Systems Management Invited Session Chair: Javad Lavaei, University of California-Berkeley, United States, lavaei@berkeley.edu 1 - Promises of Conic Relaxations in Optimal Transmission Switching of Power Systems Salar Fattahi, University of California-Berkeley, Berkeley, CA, United States, fattahi@berkeley.edu, Javad Lavaei, Alper Atamturk In this talk, we consider the problem of optimal transmission switching (OTS) in power systems. The goal is to identify a topology of the power grid that minimizes the cost of the system operation while satisfying the operational constraints. In the first part of the talk, we study the difficulty of finding a strong MILP formulation of the OTS problem. Then, we propose a convex conic relaxation of the problem based on a semidefinite program (SDP). Strong valid inequalities are proposed to strengthen the SDP relaxation by multiplying different linear constraints and then convexifying them in a lifted space. We extensively evaluate the performance of the proposed method on IEEE benchmarks systems. 2 - Opportunity for Strategic Behavior in Electricity Distribution Network Marginal Cost Based Markets Fatma Selin Yanikara, Boston University, 15 St. Mary’s Street, Room 140, Boston, MA, 02215, United States, yanikara@bu.edu, Michael C. Caramanis Transmission & Distribution (T&D) network day-ahead market co-clearing of energy and reserves must rely on Distributed Algorithms that discover Nash Equilibria of participating Distributed Energy Resources (DERs).Given the required individual DER self-scheduling - with or without the involvement of aggregators — strategic behavior may be possible under certain network information availability conditions available to DER market participants and the Network Operator. We examine the theoretical possibility for strategic behavior and quantify possible deviations from social welfare on a small T&D network. 3 - Error Bounds on Power Flow Linearizations: A Convex Relaxation Approach Daniel K. Molzahn, Argonne National Laboratory, 1111 S. Wabash Ave., Apt 1507, Chicago, IL, 60605, United States, dan.molzahn@gmail.com, Krishnamurthy Dvijotham The power flow equations relate the voltages and power injections in an electric power system. The nonlinearity of these equations results in algorithmic and theoretical challenges, including non-convex feasible spaces. Accordingly, many practical approaches for solving power system optimization and control problems employ linearizations of the power flow equations. By leveraging developments in convex relaxation techniques, this presentation describes a method for bounding the worst-case errors resulting from power flow linearizations over a specified range of operational conditions. MC25

342E Recent Development on Revenue Management and Pricing Sponsored: Revenue Management & Pricing Sponsored Session Chair: Ruxian Wang, The Johns Hopkins Carey Business School, Baltimore, MD, 21202, United States, ruxian.wang@jhu.edu 1 - New Bounds for Assortment Optimization under the Nested Logit Model Sumit Kunnumkal, Queen’s University, Kingston, ON, 500032, Canada, sk162@queensu.ca, Huseyin Topaloglu We consider the assortment optimization problem under the nested logit choice model. We establish new bounds on the quality of revenue ordered assortments. 2 - Is Ignorance Bliss? Demand Learning in Dynamic Pricing with Limited Inventory Chengyu Wu, Duke University, 877 Louise Circle, Durham, NC, 27705, United States, chengyu.wu@duke.edu, Li Chen We examine the value of Bayesian demand learning in the context of a dynamic pricing problem with inventory scarcity. An interesting and counter-intuitive numerical result is discovered where a no-learning system may outperform the corresponding Bayesian learning system. We perform analytical analysis to shed light on this observation. 3 - Omni Channel Assortment Pricing Shivaram Subramanian, IBM.Research, Yorktown Heights, NY, United States, subshiva@us.ibm.com, Pavithra Harsha, Markus Ettl We analyze certain large-scale constrained optimization problems that arise in the retail industry in the context of pricing product assortments sold across multiple sales channels and locations. The problem becomes challenging when we consider cross-product and cross-channel demand interactions. We derive an effective global optimization approach and present computational results. 4 - Pricing Ancillary Service Subscriptions Ruxian Wang, The Johns Hopkins Carey Business School, 100 International Dr, Baltimore, MD, 21202, United States, ruxian.wang@jhu.edu, Maqbool Dada, Ozge Sahin We investigate customer choice behavior in the presence of main products, ancillary services with options of pay-per-use and subscription, as well as the outside option with heterogeneous customers. 342F Spatial Decision Analysis Sponsored: Decision Analysis Sponsored Session Chair: Valentina Ferretti, London School of Economics and Political Science, London, United Kingdom, v.ferretti@lse.ac.uk 1 - Preferences in Spatial Decision Making Decision maker preferences for spatial decisions can be challenging both to model and to assess. When outcomes occur over a geographic space, the decision maker must consider not only uncertainty, risk, relative preferences for different levels of an attribute, and tradeoffs between attributes, but also the geographic areas in which each attribute level is realized. We provide several representation theorems ensuring the existence of value and utility functions for spatial decisions, explore a range of viable elicitation techniques, and demonstrate the concepts with simple examples. 2 - Sustainability of Potential Areas for Deep Geothermal Energy Systems: A Preliminary Application to Switzerland Matteo Spada, Paul Scherrer Institut, OHSA/D19, Villigen PSI, 5232, Switzerland, matteo.spada@psi.ch, Marco Cinelli, Peter Burgherr Deep geothermal energy (DGE) is a renewable technology with a secure base- load, which is gaining interest among stakeholders due to its theoretically large resource potential. However, as for other geoenergy systems, the feasibility of DGE is strongly dependent on local conditions such as, for example, geological/geophysical ones. This study presents a methodology to assess the sustainability of potential areas of interest for DGE. The proposed method is based on outranking approaches combining explicit (e.g., heat flux) and calculated (e.g., risk indicators) criteria for specific a priori defined capacity plants. The results will be presented for a case study of Switzerland. Jay Simon, American University, 4400 Massachusetts Avenue, NW, Washington, DC, 20016, United States, jaysimon@american.edu, L. Robin Keller MC24

211