Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WB47

2 - Integrating Optimized Dispatch with a Concentrated Solar Power Maintenance Simulation Model Jesse Wales, Colorado School of Mines, Golden, CO, 80401, United States, Alexander Zolan, William Hamilton, Alexandra M. Newman, Robert Braun Concentrating Solar Power (CSP) is an emerging large-scale renewable technology that uses sunlight to heat molten salt, which is stored and can be converted to power using heat exchangers and turbines. We describe a maintenance, failure and repair simulation model that interfaces with an optimization dispatch model for a CSP production plant. We optimize an energy dispatch strategy to maximize profit. We simulate planned and unplanned down- times based on dispatch decisions from the optimization model to quantify the wear-and-tear impacts from operations. 3 - Stochastic Optimization Formulations for Wind Turbine Power Maximization and Extreme Load Mitigation Yankai Cao, University of Wisconsin-Madison, 2009 Engineering Hall, 1415 Engineering Drive, Madison, WI, United States, Victor Zavala We propose stochastic programming formulations to enforce mechanical load requirements in wind turbine controller design procedures. The formulations use a probabilistic constraint that captures the long-term probability of exceeding an extreme load threshold. We use the formulations to find design parameters for pitch angle and torque controllers that maximize power output while constraining long-term extreme loads. The proposed formulations can be cast as large-scale (but structured) nonlinear programming problems (NLPs) that contain up to 7.5 million variables and constraints, and can be solved in less than 1.5 hours on a multi-core computer using existing optimization tools. 4 - Real-time Optimization of Chillers with Thermal Energy Storage and Variable Electricity Rates Landen Blackburn, University of Utah, Salt lake City, UT, United States, John Hedengren, Kody M. Powell As more renewable energy is integrated into the power grid, it is increasingly important to exploit variable electricity pricing structures to minimize commercial utility costs. Using data from one of the University of Utah’s chiller plants, equipped with a thermal energy storage tank, a novel technique is proposed to optimize this system in real time, formulated as a Mixed-integer Quadratic Programming problem. Future demand is predicted with reasonable accuracy using historic data and current weather forecasts, and the optimized result can easily be calculated in real time. The approach for solving this optimization problem proves very successful when compared to other standard algorithms. n WB47 North Bldg 229A Sustainable Operations & Development Contributed Session Chair: Feiyi Yan, Northwestern Polytechnical University, 127 West Youyi Road, Xi’an, 0, China 1 - Eco-investments: Eco-efficiency versus Eco-innovation Arda Yenipazarli, Associate Professor of Operations Management, Georgia Southern University, 2224 College Of Business, Department of Logistics & Supply Chain Mgnt, Statesboro, GA, 30460, United States We examine environmental investment efforts and whether they should be channeled towards cutting production costs through gains in process-oriented resource productivity (eco-efficiency) or product-oriented eco-differentiation that lowers customers’ water/energy/fuel use and related costs (eco-innovation), benefits that justify price premiums, and drive increased market share. While eco- efficiency is easy to pursue with little expertise and capital investment, eco-innovation requires the firm to deal with challenges of managing technology and time-to-market uncertainty. 2 - The Impact of Autonomous Vehicles on Traffic Congestion: Is Pooling the Solution to Avoiding Traffic Gridlock? Sergey Naumov, MIT, Cambridge, MA, 02142, United States We study the impact of autonomous vehicles (AVs) and ride sharing on traffic volume and road congestion. We develop a dynamic model of travel demand and demonstrate that in the absence of other regulations, the more effective AVs and pooling are at reducing congestion, the stronger is the demand response, offsetting or possibly overwhelming the assumed benefits of AVs. We show that for policies to address the negative externalities of driving effectively, they need to reduce, not increase, the relative attractiveness of driving. Our results have important implications for the design of urban transportation policies.

3 - Time Value of Different Information Dimensions on Environmental Technology Adoption Shadi Goodarzi, Assistant Professor of Operations Management, California State University Fullerton, 800 N. State College Blvd, Fullerton, CA, 92618, United States, Sam Aflaki, Andrea Masini We empirically examine the time-dependent impact of different types and channels of information on both the intention to adopt an environmental technology system and the actual adoption decision. The findings have important implications for both policy-making and for technology manufacturing companies that need to optimize their marketing strategy and distribution. 4 - Managing Funds in Non-profits: Advertisements, Fundraising and Program Expenses Nazli Turken, Assistant Professor, Cleveland State University, Cleveland, OH, 44107, United States, Janice E. Carrillo, Anand Paul In this paper, we study how non-profits allocate their funds to advertisements, fundraising activities and program expenses to maximize their return on investment. We evaluate the direct and indirect impact of spending on a non- profit’s reputation and donations, respectively. 5 - An Improved Aggregation Method for Evaluating the Performance of Serial Lines with Unreliable Machines and Finite Buffers Many analytical methods have been developed to evaluate performance of serial lines with unreliable machines and finite buffers. However, these methods cannot perform well in some cases where the machine efficiency is not uniform allocation. To solve this problem, an improved aggregation method that can perform well even in those cases is proposed. First, we demonstrate that the existence of sub-optimal machines causes the lower accuracy of the previous aggregation method. Then, an improved aggregation method is proposed by eliminating the deteriorating effect of sub-optimal machines. This method will help production managers to scientifically evaluate serial lines with higher accuracy. n WB48 North Bldg 229B Joint Session, ENRE/DM: Sustainability Topics in Urban Environments Sponsored: Energy, Natural Res & the Environment Environment & Sustainability Sponsored Session Chair: Victoria C. P. Chen, The University of Texas at Arlington, Arlington, TX, 76019-0017, United States 1 - Project-level Air Quality Assessment for Active Infrastructure Stephen Mattingly, Associate Professor, University of Texas at Arlington, Box 19308, Arlington, TX, 76019-0308, United States, Ziaur Rahman, Colleen Casey Motor vehicle exhaust emissions represent the single largest source of regional air pollution in urban areas. Outdoor physical activity near heavy-traffic is associated with symptoms of respiratory dysfunction, cardiopulmonary disease and even mortality from stroke. At present, project-level air pollution concentration standards do not exist, so the development of air quality performance measures and standards for physical activity will help decide future activity paths and infrastructure priorities. This research creates performance measures for evaluating the relationship between air quality and the physical characteristics of walking and cycling routes. 2 - Modeling and Stochastic Bounds on Multi-agent Demand Response Alireza Fallahi, TX, United States, Jay Michael Rosenberger, Vicoria C.P, Chen, Wei-Jen Lee The renewable energy revolution will substantially alter the future energy. With increasing penetration of stochastic renewable energies, the power distributors must respond to an extremely dynamic, complex, and challenging real-time supply-demand balancing problem, which lead to a dramatically different energy market, both operationally and fundamentally in the near future. This research is designed to bridge the knowledge gap. The mathematical model is proposed and bounds of the optimal operational cost are evaluated. Fei-Yi Yan, Northwestern Polytechnical University, 127 West Youyi Road, Xi’an, China, Jun-Qiang Wang

458