2015 Informs Annual Meeting

SC57

INFORMS Philadelphia – 2015

SC57 57-Room 109B, CC Energy Technology, Climate Change, and Uncertainty Sponsor: ENRE – Energy II – Other (e.g., Policy, Natural Gas, Climate Change) Sponsored Session Chair: Erin Baker, University of Massachusetts, MIE Department, 220 ELAB, Amherst, MA, United States of America, edbaker@ecs.umass.edu 1 - Equilibrium vs. Optimality: Trading in Renewable Energy Certificates Ekundayo Shittu, George Washington University, Washington, DC, United States of America, eshittu@email.gwu.edu, Linus Nyiwul We propose the harmonization of independent renewable energy credit markets, and study their impacts on a firm’s energy technology choice and capacity decisions. The industry is struggling with this issue right now, and we inform this policy debate by comparing market mechanisms in which each participant independently maximizes their targets. We find that while there are optimal market conditions, the equilibrium is not only unstable, the overall efficiency gains are not always positive either. 2 - Managing Climate Risks with Carbon Mitigation: A Stochastic Programming Approach with Merge Delavane Diaz, Stanford University, Huang Engineering Center, Stanford, CA, 94305, United States of America, delavane@stanford.edu, Geoffrey Blanford Carbon policy is fundamentally about risk management – balancing the costs of reducing emissions and the benefits of avoided climate change, both of which are uncertain due to incomplete scientific understanding and complex interactions. This paper presents a framework for decisionmaking under uncertainty in MERGE, examining optimal carbon mitigation given uncertainty about the physical climate system and climate damages. This work provides insight into managing downside risks of climate change. 3 - An Approximate Dynamic Programming Algorithm for Unit We present a novel formulation of a stochastic unit commitment model including energy storage using approximate dynamic programming. We demonstrate that the non-linear dynamics of energy storage lead to different optimal strategies for using storage as compared with the typical linear formulation used in most UC models. 4 - An Approach to Deep Uncertainty in Climate Change: Robust Portfolio Decision Analysis Erin Baker, University of Massachusetts, MIE Department, Commitment with Energy Storage Mort Webster, mdw18@psu.edu We advance the concept of Robust Portfolio Decision Analysis and apply it to analyzing public energy technology R&D portfolios in response to climate change. We consider 3 sets of expert elicitations over 5 energy technologies. We identify technology projects that are in all, none, or some of the non-dominated portfolios, where non-dominated is defined in terms of multiple priors. We discuss the implications for value of information and for generating new alternatives with high option value. SC58 58-Room 110A, CC Resiliency and Reliability Optimization of Electric Power Systems Sponsor: ENRE – Energy I – Electricity Sponsored Session Chair: Frank Felder, Associate Research Professor, Rutgers University, 33 Livingston Ave, New Brunswick, NJ, 08901, United States of America, ffelder@rci.rutgers.edu Co-Chair: David Coit, Professor, Rutgers University, coit@rci.rutgers.edu 1 - Long-Term Mitigation for Improved Restoration in Power Networks Emily Heath, Graduate Student, Rensselaer Polytechnic Institute, 110 8th St., Troy, NY, 12180, United States of America, heathe@rpi.edu, Thomas Sharkey, John Mitchell This research looks at how the best mitigation plan can be selected for a power network using a ranking and selection procedure. The power system is modeled using the direct current (DC) model, and a performance measure is developed to 220 ELAB, Amherst, MA, United States of America, edbaker@ecs.umass.edu, Valentina Bosetti, Ahti Salo

measure how a mitigation plan can contribute to the rapid restoration of the network following a disruption. We discuss the computational challenges of using the DC model, and compare results using a flow-based model on the same network. 2 - Combined Natural Gas and Electric System Operation with Wind Energy Dan Hu, Iowa State University, 3004 Black Engineering Bldg, Ames, IA, United States of America, danhu@iastate.edu, Sarah Ryan We formulate a model of a combined natural gas and electric power system including wind energy. A two-stage stochastic programming model for day-ahead scheduling is proposed with uncertainty in wind power production. Joint optimization of gas delivery and electricity production, with the ability to store natural gas, help to maintain equilibrium in the combined system while meeting demand with high reliability and low cost. 3 - An Adjustable Robust Optimization Approach to Provision of Interruptible Load Qi Zhang, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, qi.zhang13@gmail.com, Michael F. Morari, Ignacio E. Grossmann, Jose M. Pinto, Arul Sundaramoorthy In modern electricity markets, large electricity consumers can sell operating reserve by providing capacities to reduce their electricity load upon request. Providing such interruptible load can be very lucrative; however, one does not know in advance when load reduction will actually be requested. In this work, an adjustable robust optimization approach is applied to model this uncertainty, using affine decision rules that allow recourse decisions in the resulting scheduling problem. SC59 59-Room 110B, CC Just the Facts: Empirical Patterns in Strategy Cluster: Strategy Science Invited Session Chair: Myles Shaver, University of Minnesota, 321-19th Ave S, Suite 3-365, Minneapolis, MN, 55455, United States of America, Mshaver@umn.edu 1 - How Competition Affects the Governance of R&D Projects: Evidence from Biotechnology Clinical Trials Mazhar Islam, Drexel University in Philadelphia, PA mui27@drexel.edu Although almost all biotechnology firms participate in R&D alliances, we highlight that when one looks at a more micro-level of analysis – drug compounds within a therapeutic area – the majority of projects are done internally. Using a unique data set of clinical trials in 24 therapeutic areas in the U.S. biotechnology industry between 1996 and 2008, we show that biotechnology firms prefer internal organization absent competition from other biotechnology firms in the therapeutic area. With greater competition, we observe that these firms are more likely to utilize non-equity alliances compared to internal development - presumably to speed time to market within a competitive arena. We present two contingencies that aid in identifying the mechanism underlying this empirical finding – scope of applicability of the drug compound and the biotechnology firm’s previous success with drug development projects. 2 - Innovation and Competition among Different Size FIrms Siddharth Sharma, PhD Candidate, Strategic Management, Robert H. Smith School of Business, University of Maryland, MD, United States of America, siddharth@rhsmith.umd.edu, Wilbur Chung We examine the Consumer Electronics trade show (CES) as a microcosm of competitive interaction among different size firms. In this dense space, firms seek to position their booths to maximize exposure during this punctuated event. While industry heavy weights occupy key spots and little known ones are in the periphery, we still observe that firms of quite different sizes can be neighbors. We expect smaller firms armed with an innovation to seek out larger firms. We develop a simulation model with different size firms that have differing probability and value of innovations. Once their innovation draw is known, firms chose where to locate on a two-dimensional space with heterogeneous demand and look to maximize their demand. Firms compete by locating adjacent to others to capture some of their neighbors’ demand. But locating with others can also generate externalities – agglomeration economies – that may offset competition. We compare the simulation’s predictions versus actual booth locations. The setting and resulting simulation provide insights into the competitive dynamics underlying industry evolution.

112