2016 INFORMS Annual Meeting Program

SB60

INFORMS Nashville – 2016

SB58 Music Row 6- Omni Energy II Contributed Session

SB59 Cumberland 1- Omni TSL Keynote Address Keynote Session

Chair: Alberto J Lamadrid, Assistant Professor, Lehigh University, 621 Taylor Street, R451, Bethlehem, PA, 18015-3120, United States, ajlamadrid@ieee.org 1 - Optimal Transmission Line Switching Under Geomagnetic Disturbances Mowen Lu, Clemson University, Industrial Engineering Department, 801 7th St. Apt 21B, Los Alamos, NM, 87544, United States, mlu87@g.clemson.edu, Russell Bent, Scott Backhaus, Harsha Nagarajan, Emre Yamangil Geomagnetically-induced current (GIC) flows induced by geomagnetic disturbances (GMDs) can cause transformer hotspot heating and reactive power losses that can severely impact power grid reliability. We present an optimal transmission line switching model for a power grid experiencing a GMD. We show how convex quadratic relaxations and improved bilinear function relaxations can be used to obtain tight lower bounds. A case study based on a modified single area of the IEEE RTS-96 system demonstrates our model’s operating decisions with respect to GMD direction. 2 - Management Of Electric Vehicles Participation In Smart Grids For Demand Response Nasim Nezamoddini, Ms., State University of New York - Binghamton, 222 Main Street, Apt 1, Binghamton, NY, 13905, United States, nasim.nezam@gmail.com Electric vehicles (EVs) are potential distributed energy resources that support both the grid-to-vehicle and vehicle-to-grid modes in the smart grid. Their participation in the time-based and incentive-based demand response (DR) programs helps improve the stability and reduce the potential risks to the grid. This research proposes a novel stochastic model from the Independent System Operator’s perspective for risk management and participation planning of EVs in the smart grid for DR. The effectiveness of the model in response to various settings such as the area type (residential, commercial, and industrial), the EV penetration level, and the risk level has been investigated. 3 - A Computationally Efficient Branch-and-cut Method For Robust Unit Commitment Unit commitment seeks the most cost effective generator commitment decisions to meet net load while satisfying operational constraints. Robust optimization is a widely studied approach for optimizing under uncertainty in the load less variable generation. We adopt a branch-and-cut approach to solve a Benders decomposition for the robust formulation with different forms of uncertainty sets. Its improved efficiency relative to the naïve Benders algorithm is demonstrated with numerical experiments. 4 - Modeling And Computation Of Reliable Grid Expansion Bo Zeng, University of Pittsburgh, Department of Industrial Engineering, Pittsburgh, PA, 15260, United States, bzeng@pitt.edu, Hossein Haghighat Different from existing capacity expansion model, we consider the non- cooperative market clearing results in bilevel capacity expansion scheme. Numerical results will be presented to demonstrate this novel scheme in practice. 5 - Interdependencies In The Communications And Electrical Networks Alberto J Lamadrid, Assistant Professor, Lehigh University, 621 Taylor Street, R451, Bethlehem, PA, 18015-3120, United States, ajlamadrid@ieee.org, Basel Alnajjab, Lawrence V. Snyder, Rick Blum, Shalinee Kishore Managing an electric grid in a reliable and economic manner requires collecting information for grid operators. Part of this information is collected through sensors and a supervisory control and data acquisition (SCADA) system, as well as from data provided by market players in the market clearing process. Therefore, the operation of the system requires three interdependent networks: the electric grid, the communication network and the control network. This paper presents a model that explicitly connects these three networks, considering the probabilistic nature of possible disruptions and changes in any of the networks. Narges Kazemzadeh, Iowa State University, 3004 black engineering, Ames, IA, 50011-2164, United States, narges.kazemzadeh@gmail.com, Sarah M Ryan

Chair: Stephen D Boyles, The University of Texas at Austin, 301 E. Dean Keeton St. Stop C1761, Austin, TX, 78712, United States, sboyles@gmail.com 1 - City Logistics: Challenges And Opportunities Martin W P Savelsbergh, Georgia Institute of Technology, School of Industrial and Systems Engineering, Atlanta, GA, 30332-0205, United States, martin.savelsbergh@isye.gatech.edu Tom Van Woensel Today, around 54% of the world’s population lives in urban areas. By 2050, this share is expected to go up significantly. As a result, city logistics, which focuses on the efficient and effective transportation of goods in urban areas while taking into account the negative effects on congestion, safety, and environment, is critical to ensuring continued quality of life in cities. We review and discuss a variety of current and anticipated challenges and opportunities of city logistics. We hope this helps shaping an appropriate research agenda and stimulates more researchers to enter this exciting field. SB60 Cumberland 2- Omni Advances in Traffic Equilibrium and Network Loading Models Sponsored: TSL, Urban Transportation Sponsored Session Chair: Yang Liu, National University of Singapore, National University of Singapore, Singapore, 00000, Singapore, iseliuy@nus.edu.sg 1 - Time-based Equilibrium For Staggered Bottleneck Congestion With Heterogeneous Commuters Yang Liu, National University of Singapore, iseliuy@nus.edu.sg This paper develops a time-based dynamic user equilibrium (DUE) model with heterogeneous users under the flexible work hour scheme. We first prove that, given the identical work start time, the equivalent traffic assignment can be formulated as a symmetric and convex problem for time-based DUE, which admits unique solution. At short-run DUE, i.e., the staggered work hour scheme is predetermined, DUE is formulated as variational inequality, and the unique solution cannot be guaranteed. At long-run DUE, where commuters have the flexibility to choose work start time, the equivalent convex optimization is formulated. We prove that both DUE solution and commute cost are uniquely determined. 2 - Price Of Satisficing: Bounding The Performance Of Boundedly Rational User Equilibrium Mahdi Takalloo, University of South Florida, Tampa, FL, United States, mtakalloo@mail.usf.edu Changhyun Kwon Due to the lack of full information and personal preferences, network users are mostly satisficing decision makers instead of perfectly rational decision makers. This satisficing behavior results in a traffic pattern, which attains a boundedly rational user equilibrium (BRUE) instead of perfectly rational equilibrium (PRUE). This research uses a perception error model to describe BRUE, and also quantifies how bad the system total travel delay in BRUE can be compare to PRUE. 3 - Path Based Formulations For Generalized Bounded Rationality And Multi-commodity Network Design Changhyun Kwon, University of South Florida, 4202 E. Fowler Ave. ENB118, Tampa, FL, 33620, United States, chkwon@usf.edu Longsheng Sun, Mark H Karwan The route choices of network users could be uncertain when they are satisficing with sub-optimal routes or there are errors in each individual’s perception of the route’s utility. By modeling perception error, the notion of generalized bounded rationality was recently proposed, providing a unified framework for the two sources of uncertainty in route choices. In this paper, we design path based formulations for route choice uncertainty represented by generalized bounded rationality. We show how to incorporate them into multi-commodity network design problems.

63