Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WA49

3 - Valuation of Long-duration Grid Energy Storage using Production Cost Models Jiazi Zhang, Postdoctoral Researcher - Electrical Engineering, National Renewable Energy Laboratory, 15013 Denver West Parkway, Golden, CO, 80401, United States, Joshua Eichman, Omar Jose Guerra Fernandez The integration of high shares of variable renewable energies raises challenges for the reliability and cost-effectiveness of power systems. Long-duration storage is one of the most promising options to solve these problems. In this study, the system-level impacts of long-duration storage on the Western Interconnect (WI) power system will be explored. The operations of future WI with 50%, 70%, and 90% renewable penetration are simulated using a production cost modeling tool - PLEXOS. The net system benefits of different storage and demand response technologies are assessed including pumped hydro, compressed air energy storage, flow batteries, power-to-gas, and power-to-gas-to-power. 4 - Electric Generation Investment under Uncertainty: Applying Decision Analysis to a Complex Electric Power Systems Model David Young, EPRI, Palo Alto, CA, 94304, United States, Nidhi Santen We adapt the deterministic US-REGEN inter-temporal capacity expansion model to approximate a stochastic model by wrapping it within a decision tree framework. We apply this framework to study how the anticipation of future climate policy or natural gas price shocks in the U.S. impacts generation investment decisions ex-ante. We find investing in new wind/solar or gas-fired units is generally robust to a range of futures, but investing in new nuclear units or units with carbon capture and storage prior to the shock may lead to high regretted costs. 5 - Performance Evaluation in Distributed Photovoltaic Systems Using Data Envelopment Analysis Adiel Teixeira De Almeida Filho, Universidade Federal de Pernambuco, Caixa Postal 7471, Av. General San Martin, 1083, Recife-PE, 50.630-971, Brazil, lvaro Cavalcanti, Francisco de Assis dos Santos Neves, Gustavo Medeiros Azevedo This paper uses the Data Envelopment Analysis (DEA) technique to verify the performance of micro- and mini-generation systems in order to reveal the best practices that could be observed so that these may be replicated in order to justify the increase in photovoltaic generation in this type of system that will be distributed and managed by the population. During the analysis, data were collected on 123 micro- and mini-generation units distributed in an area of 98.5 km , in the Northeast region of Brazil. 6 - A Lattice Method for Jump-diffusion Process Applied to Transmission Expansion Investments under Demand and Distributed Generation Uncertainties Fikri Kucuksayacigil, Iowa State University, Ames, IA, 50010, United States, K. Jo Min In recent years, decision makers of expansion investments in transmission sector have faced critical uncertainties such as growth of demand for electricity and installation of distributed generations (DGs) owned by customers. Power lines should be expanded strategically as installation of DGs may create stranded costs for transmission owners as they meet a portion of local demand. We develop a computationally efficient lattice method to model uncertainties in the form of geometric Brownian motion and compound Poisson process. We propose a real options framework which quantifies the value of expansion investments. Proposed framework is demonstrated on a hypothetical example. n WA49 North Bldg 230 Practice- Game Theory II Contributed Session Chair: Patanjal Kumar, Indian Institute of Management Rohtak, MDU Campus, Everest Hostel, Rohtak, 124001, India 1 - A Decentralized Algorithm for Finding Nash Equilibrium under Imperfect and Incomplete Information Kyuree Ahn, KAIST, Daejeon, Korea, Republic of, Jinkyoo Park In this study, we propose a nonparametric decentralized algorithm to find a Nash equilibrium of a repeated game with imperfect and incomplete information. At each iteration of the game, each player executes his own action and observes the reward and other players’ executed actions (strategies). Each player then updates his belief both on his own reward function and on other players’ joint strategy using Gaussian Process Regression and Log Gaussian Cox Process, respectively. As each player best responds to their beliefs that is being updated, the joint of individually updated strategy converges to a Nash equilibrium in a distributed way.

2 - Socially Differentiated Products: A Game Theoretic Approach Hannan Sadjady Naeeni, University of Houston, Houston, TX, United States, Powell Robinson We propose a game-theoretic duopoly model to identify equilibria configurations where one, none, or both of the firms offer socially responsible (SR) products. Our analysis shows how firms’ commitment to socially responsible products depends on the marginal cost differences, market growth, and changes in consumer’s perception of brand. 3 - Information Aggregation via Rational Learning in Imperfect Information Routing Games Manxi Wu, Massachusetts Institute of Technology, Cambridge, MA, 02142, United States, Saurabh Amin, Asuman Ozdaglar We study rational learning dynamics in traffic routing games when travelers have imperfect information about an unknown network state. In each stage, travelers myopically choose routes based on their current belief distribution, and noisy realizations of travel costs on the chosen routes become publicly available. We show that rational learning can lead to a self-confirming equilibrium in which average traveler cost is higher than the cost under full knowledge of the state. We specialize our results in situations in which self-confirming equilibrium does not exhibit such informational inefficiency, i.e., travelers make route choice as if they have complete information about the state. 4 - Using Auditing and Commitment for Engaging a Supplier for Supply Chain Sustainability Hossein Rikhtehgar Berenji, PhD Candidate, University of Oregon, Lundquist College of Business, 1208 University of Oregon, Eugene, OR, 97403-1208, United States, Nagesh N. Murthy, Zhibin Yang We model a buyer auditing a supplier for sustainable business practices. We investigate the effect of buyer’s upfront commitment to price and quantity on supplier’s compliance with the code of conduct and buyer’s auditing effort and supply chain sustainability. We analyze implications of raising the standard of Patanjal Kumar, Indian Institute of Management Rohtak, MDU Campus, Everest Hostel, Rohtak, 124001, India, Peeyush Pandey, Gourav Dwivedi This paper presents a novel approach to coordinate green and socially responsible supply chain using contracts. We compare the traditional and virtually integrated form of an organization using game theoretic approach. Through numerical examples, we demonstrate the effectiveness of our models. Our results reveal that virtual organization performs better than the traditional form of an organization under specific conditions. n WA50 North Bldg 231A Practice- Retail Management II Contributed Session Chair: Wenli Peng, Katholieke Universiteit Leuven, Voie du Roman Pays 34 bte L1.03.01, Louvain la Neuve, 1348, Belgium 1 - Win the Buy Box - Reselling or Marketplace Hao Su, University of Maryland-College Park, 3330 Van Munching Hall, College Park, MD, 20742, United States, Laharish Guntuka, Martin E. Dresner This paper examines an internet retailer’s decision to directly sell a product. The internet retailer offers customers both direct sales and sales through a marketplace. Using empirical analysis, we determine product characteristics that predict the internet retailer’s decision to directly sell a product. We hypothesize that the internet retailer is more likely to sell (1) popular (vs. unpopular) products and (2) the products that have large (vs. limited) spillovers onto other products. Findings provide implications to marketplace sellers seeking to avoid direct competition with the internet retailer. 2 - An Investigation of Multi-staged, Dual-pronged Order Fulfillment in E-commerce Marketplace Hyun Seok (Huck) Lee, Assistant Professor, College of Business, Oregon State University, 3007 NW Morning Glory Drive, Corvallis, OR, 97330, United States, Yusoon Kim, Junbo Son Increasingly large e-commerce marketplaces bring order fulfillment function in- house. As a result, we observe two internal fulfillment systemsùconventional fulfillment by individual merchants (FBM) and emerging fulfillment by e- commerce platform (FBP)ùand they compete internally over efficiency. In this research, based on real-world e-commerce data on transactions and logistics, we compare the two internal fulfillment systems through uncovering distinctive mechanisms driving different operational and logistical decisions. By doing so, we suggest contingent conditions for moving towards the FBP option. code of conduct on sustainability and financial performance. 5 - Impact of Virtual Organization in Sustainable Supply Chain Contracts

428