Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

MC47

the tradeoff between short-term genetic gains and long-term growth potential. Our contribution is to define a new look-ahead metric for assessing a selection decision, which evaluates the probability to achieve both genetic diversity and breeding deadline. Moreover, we propose a heuristic algorithm to find an optimal selection decision with respect to the new metric. Our new selection method is

n MC47 North Bldg 229A A Guide to Optimization Based Multi-Period Planning Emerging Topic Session Chair: Esma S. Gel, Arizona State University, School of Computing, Informatics and, Decision Systems Engineering, Tempe, AZ, 85287- 8809, United States 1 - A Guide to Optimization Based Multi-Period Planning Linus Schrage, LINDO Systems, Chicago, IL, 60637, United States Many organizations use multi-period planning models that involve optimization to decide things like the best production or investment levels in multiple periods into the future. There are a wide variety of features a user would like to have in such models. How those features are represented affects both the usefulness of the results and the solvability of these models as you add more periods to the model, or add more products, or in general, increase the detail. This tutorial describes how to best represent some important features that are common to most long range planning models. (a) Planning horizon length. (b) Ending conditions. The final period of the planning model frequently needs special treatment. In some situations you may be able to actually use an infinite horizon plan. (c) Period length. (d) Uncertainty. What is the best way of representing it? Variance, downside risk, Value-at-Risk, a utility function of some sort? (e) “Nervousness” and “sliding scheduling”. Most planning models are used in a “rolling or “sliding fashion, e.g., solve a 12 period model this month, implement the first period, and then next month slide things forward and repeat. When this is done, “nervousness may be a problem, i.e., the plan made in February for what to do in June may differ substantially from what the plan published in January suggested for June. (f) Changeover, startup and shutdown costs. (g) Precedence constraints. (h) Scarce resource constraints. (i) Taxes. These can be important in some planning models. How these are properly calculated, or at least approximated, in an optimization model can be a challenge in the presence of features such as depreciation and choice of FIFO vs. LIFO inventory valuation. (j) Nonlinearities. n MC48 North Bldg 229B Modeling Sustainability and Energy Sponsored: Energy, Natural Res & the Environment Environment & Sustainability Sponsored Session Chair: Zana Cranmer, Bentley University, 175 Forest Street, Waltham, MA, 02452, United States 1 - Effect of Nuclear Waste Management Costs on the Economic Viability of Nuclear Energy as a Low Carbon Energy Option Robert Barron, Western New England University, Lawrence, KS, 66047, United States, Mary C. Hill We consider how a more realistic treatment of the nuclear waste disposal than has been used in previous studies could affect the viability of nuclear power in the context of integrated assessments of climate change. Our results suggest that the optimism reflected in previous works is fragile: More realistic nuclear waste management cost models and uncertainty-appropriate inter-generational discount rates produce many more scenarios in which nuclear waste management costs are higher than previously assumed, and nuclear energy’s economic attractiveness as a low carbon energy option is lower than earlier works suggested. 2 - Competition Between Subsistence and Organized Retailers Jiwen Ge, Postdoc, Tuck Business School, Woodbury 005, 100 Tuck Mall, Hanover, NH, 03755, United States, Brian Tomlin, Jan C. Fransoo Subsistence retailers are independent small stores through which store owners provide for their family’s subsistence needs. We study the competition between subsistence retailers and an organized retailer in an emerging market setting. For a model without consideration of the supplying manufacturer, we show the entry of the organized retailer could be detrimental to subsistence retailers. However, if we include the manufacturer in our analysis, we show that he is incentivized to set a favorable price so that subsistence retailers can survive sustainably. 3 - Look-ahead Selection Strategy for Genomic Selection: Optimizingselection and Pairing Strategies with a Time-dependent Approach Guiping Hu, Iowa State University, 3014 Black Engineering, Ames, IA, 50011, United States, Saba Moeinizade, Lizhi Wang, Patrick Schnable Genomic selection (GS) techniques allow breeders to make decisions with the genotypic data at an early stage. A major limitation of existing GS approaches is

outperforming the other selection methods in the literature. 4 - Spatially General Probabilistic Prediction Model for Wind Power Output

Kristen Schell, Postdoctoral Research Fellow, Polytechnique Montreal, 2900 Edouard Montpetit Blvd, Montreal, QC, H3T 1J4, Canada, Andrea Staid, Seth Guikema Government incentives, such as production tax credits, have successfully spurred investment in wind power. Yet, this wind power is increasingly curtailed - deemed unusable by the power system - at astonishing rates. This reality highlights a knowledge gap in how to site wind farms that best contribute to power system resource adequacy. Using meteorological and historical wind power output data from several wind farms across different terrains, we develop a probabilistic prediction model for wind power output. This model gives the cumulative distribution function of predicted power output at any location, allowing system planners to prioritize wind farm development in high-producing locations. 5 - The Drivers of Renewable Energy Innovation; An Empirical Analysis Saeedeh Anvari, Postdoctoral Researcher, Darden School of Business, 136 University Gardens, Charlottesville, VA, 22903, United States, Michael J. Lenox In this paper, we analyze the drivers of renewable energy innovation with the goal of discerning what drives both the rate and direction of innovative activity. We focus on public sector financing, private sector investments, and policy support for innovation as potential important drivers. Empirically, we measure innovation, using patents, across different economic regions in the US leveraging differences in regional ecosystems to tease out the impact of specific drivers. n MC49 North Bldg 230 Big Data Analytics for Higher Education Industry Contributed Session Chair: Roger R. Gung, University of Phoenix, 3842 E. Windsong Dr., Phoenix, AZ, 85048, United States 1 - Analytics Framework for Higher Education Industry Roger R. Gung, Sr. Director, Business Analytics & Operations Rese, University of Phoenix, 4025 S. Riverpoint Parkway, Phoenix, AZ, 85048, United States We present an end-to-end analytics framework for adult learning/higher education industry that covers solution methods with using big data analytics for the management of marketing, enrollment service, learning platform, academic performance and institutional finance. 2 - Improve Academic Performance by Optimizing Faculty Assignment Mei Duanmu, University of Phoenix, 4025 S. Riverpoint Parkway, Phoenix, AZ, United States, Jianhua Duanmu, Roger Gung It is difficult to evaluate teachers’ effectiveness in the educational industry, due to the non-random assignment of students to teachers. The performance for some teachers are overestimated/underestimated, because they teach the good/bad students. In order to filter out the impact of student quality difference, we developed the mixed effects model using student attributes as fixed effect and teacher impact as random effect. Separate models are developed for hundreds of courses in University of Phoenix. Based on the model outputs, teachers are then ranked by their effectiveness from top to down, which are used to guide the teacher assignment for future course sessions. Compared to random assignment, we see significant and consistent improvements in student performance by assigning teachers based on their rankings. 3 - Hadoop Big Data System for eCampus and Educational Analytics Jixiang Fang, University of Phoenix, Phoenix, AZ, United States, Jianhua Huang, Roger Gung Big data is a technique to build a data platform for collecting and managing large volume or high dimensional data. Nowadays, many companies begin to increase their information technology investment in the big data field. A well-designed big data system is reliable for data mining and decision making. Hadoop system is a popular big data framework for storing data and running applications such as statistical programming software. This data system at the University of Phoenix (UOP) has been developed in the past 8 years by the data team and it is used to collect and synchronize the data of student learning activities and faculty teaching activities that were posted on the eCampus platform. In this presentation, the big data system at UOP will be an example to illustrate the procedure for using big data.

202