Informs Annual Meeting 2017

TB78

INFORMS Houston – 2017

TB78

TB78B

381A Leveraging OR to Assess and Enhance Power Systems Resilience Sponsored: Energy, Natural Res & the Environment Electricity Sponsored Session Chair: Roshi Nateghi, rnateghi@purdue.edu 1 - Assessing Changes in Residential Electric Power Consumption Due to Emerging Changes in Climate and Technology Hua Cai, Purdue University, West Lafayette, IN, 47907, United States, huacai@purdue.edu, Roshanak Nateghi, Ruoxi Wen, Benjamin Rachunok The electricity sector - particularly the residential sector - is undergoing significant transformations due to the emerging changes in technology (e.g., distributed generation, electric vehicles), policy, and climate. However, the existing energy systems models (e.g., NEMS, MARKAL) are highly deterministic and do not adequately account for the uncertainties associated with future climate or technology adoption scenarios. Using agent-based models, we aim to estimate the residential electricity use over the next 30 years, considering not only the energy demand for space conditioning with different housing characteristics, but also the adoption and use of EV and PV at the household level. 2 - Assessing Residential Power Inadequacy Risks in the U.S. Roshanak Nateghi, Purdue University, West Lafayette, IN, 47907, United States, rnateghi@purdue.edu The US power system is increasingly vulnerable to adverse impacts of extreme weather and climate events. Supply inadequacies can result from unanticipated climate-induced shifts in demand. This talk is primarily focused on the risks associated with deviated residential demand patterns due to climate variability and change. A data-driven approach is leveraged to identify and assess the risk factors that render the residential electricity sector vulnerable in face of future climate variability and change. The proposed method, can be used by the energy professionals and regulators to design effective mechanisms to minimize supply inadequacy in face of climate variability and change. 3 - Enhancing Resilience of the Electric Power Sector after Hurricane Sandy This talk presents a framework to help a decision maker allocate resources to increase his or her organization’s resilience to a system disruption, where resilience is measured as a function of the average loss per unit time and the time needed to recover full functionality. Enhancing resilience prior to a disruption involves allocating resources from a fixed budget to reduce the value of one or both of these characteristics. The optimization model is applied to an example of increasing the resilience of an electric power network following Superstorm Sandy. 4 - Parallel Computing of Stochastic Programs with Application to Energy System Capacity Expansion Run Chen, Purdue University, 1399 Neil Armstrong Drive, Apt 7, West Lafayette, IN, 47906, United States, chen885@purdue.edu, Andrew Lu Liu Power grids’ planning and operations exhibit extreme multiscale, ranging from hourly operation to decades of planning. The linkage such decisions can be treated in a primal-dual fashion to produce multiple independent subproblems. We propose to use a primal-dual-based proximal method to design a parallel algorithm to solve such multiscale problems. Convergence can be shown for both convex and non-convex problems. We use the algorithm to demonstrate the importance to consider ramping constraints in finer time-scale in long-term capacity planning with large-scale renewable resources. 5 - Possible Improvements for Generation Reserves to Back-up Wind in the Dutch Electricity Market Robin Broder Hytowitz, Johns Hopkins University, 3400 North Charles Street, Ames 313, Baltimore, MD, 21218, United States, hytowitz@jhu.edu, Ozge Ozdemir, Paul Koutstaal, Benjamin Field Hobbs Reserve capacity is needed in an electric system in case of a contingency. With the increased penetration of renewable energy in the Netherlands, a new type of renewable contingency can require additional reserve on the system. The Netherlands currently procures reserve months in advance through long time contracts with little coordination with its neighbors. This project suggests improvements that can be made to the process of procuring, allocating, and activating reserve. A nodal model of Europe is used to analyze the improvements, employing a future scenario with high wind penetration. Comparison of different improvements will elucidate which have the most benefits to the system as a whole. Cameron MacKenzie, Iowa State University, 3029 Black Engineering, Industrial and Manufacturing Systems Eng, Ames, IA, 50011, United States, camacken@iastate.edu, Christopher Zobel

380B Automotive Contributed Session Chair: Mark Colosimo, Urban Science Applications, Inc., Detroit, MI, United States, macolosimo@urbanscience.com 1 - Location Optimization Solution in Car Auction Industry Wennian Li, Data Scientist, Cox Automotive, 18th Floor, 1001 Summit Blvd, Atlanta, GA, 30319, United States, li_wennian@hotmail.com Cox Automotive is a leading provider of products and services for automotive dealers and car buyers. The company unites 30+ brands, including AutoTrader.com, Kelley Blue Book, Manheim, NextGear Capital, and vAuto, to help transform the way people buy and sell cars. The Location Optimization project helps the sellers to determine where to sell their inventory and gain more profit/ROI without lower the sale through rate and velocity at both wholesale and retail market. We provide a package solution for the seller not only just the suggested solution to locate their inventory, but also interactive interface for the current inventory, market status and suggested solutions. 2 - Using Leading Economic Indicators to Predict Changes in Monthly Car and Light Truck Sales in the United States Market Karl Majeske, Associate Professor, Oakland Unverisity, School of Business Administration, Rochester, MI, 48309, United States, majeske2@oakland.edu This research uses economic indicators (e.g., unemployment rates, gas price, consumer confidence, ect.) as independent variables to predict changes in car and light truck sales. Each economic indicator is evaluated in a variety of ways (actual values, monthly change and annual change) to determine the best measure to relate with sales. In addition, the economic indicators are lagged over a collection of time periods to identify the time lag that best correlates it’s change to a related change in car and light truck sales. Establishing representations and time lags for a collection of economic indicators will allow determining how to identify an economic signal that future sales will change. 3 - Automobile Logistics Optimization Between Mexico and the United States Ahad Ali, Associate Professor, Lawrence Technological University, 21000 West Ten Mile Road, 21000 West Ten Mile Road, Southfield, MI, 48075, United States, aali@ltu.edu The goal of this project is to optimize the logistics of an automobile company between Mexico and the U.S. The logistics involved are the ones between the manufacturing plants of one of these companies in Mexico, how they interact between each other, and how they transport the products from one country to another. The methods that are going to be used are optimization methods and simulation. The possible outcome of this research is to optimize the logistics method one of these companies use in order to reduce costs and time. Optimization techniques and simulation will be done in order to reduce the costs and time of transferring the car parts between plants and countries. 4 - The Future of Autonomous Carsharing - A Delphi Study Katrin Merfeld, EBS.Universitaet fuer Wirtschaft und Recht, Gustav-Stresemann-Ring 3, Wiesbaden, 65189, Germany, katrin.merfeld@ebs.edu, Mark-Philipp Wilhelms, Karin Kreutzer, Sven Henkel Autonomous driving is set to disrupt the mobility landscape in the coming years. Academia predominantly examines individual aspects of this novel technology, such as enhanced traffic safety or ethical issues. As individual adoption may be hindered by financial hurdles, autonomous cars will be especially relevant for carsharing practices. We conducted a four-stage Delphi study with 40 international experts from various sectors to elicit drivers, barriers and future developments of carsharing with autonomous vehicles. We discuss our findings and draw implications for managers and further academic research. 5 - Time to Recall Decision Making Chelsey Helena Hill, Drexel University, Gerri C. LeBow Hall Room #730, 3220 Market Street, Philadelphia, PA, 19104, United States, chh35@drexel.edu, Chaojiang Wu This paper investigates the impact of the time-to-recall decision on marketing and operations performances in the automobile industry. While we find that the consumer market seems to have little response to the delay of recalls, recall effectiveness deteriorates. This motivates the creation of an analytical model to quantify the impact of this delaying effect on recall costs. Using the estimated relationship between time and effectiveness from our empirical model, we present an analytical model minimizing total recalls costs. The findings reveal that an early recall decision is primarily driven by higher liability costs, and later recall decisions are driven by the cost of sales disruption.

332