2015 Informs Annual Meeting

MD79

INFORMS Philadelphia – 2015

3 - Purchasing Postponement and SC Coordination in a Decentralized N-V Model with Stochastic Demand

3 - A Multi-period Energy-Aware Inventory Model with CVAR Constraints

Sourabh Bhattacharya, Professor, Institute of Management Technology, Hyderabad, India, 38, Cherlaguda Village Shamshabad, Hyderabad, TS, 500048, India, sbhattacharya@imthyderabad.edu.in

Niloofar Salahi, Graduate Research Assistant, Rutgers The State University of New Jersey, 96 Frelinghuysen Road, Piscataway, NJ, 08854, United States of America, niloofar.salahi@gmail.com, Mohsen Jafari A risk-averse production planning with energy efficiency consideration is introduced for an industrial process subject to stochastic demand. We present an inventory model that minimizes expected costs while maintaining performance requirements. The energy consumption is calculated using energy-performance curves specific to the type of industrial process. We show that significant cost saving is expected when adjusting the production plan according to time dependent electricity pricing schemes. 4 - Robust Optimization vs. Stochastic Programming for Electricity Generating Unit Commitment Narges Kazemzadeh, Graduate Research Assistant, Iowa State University, Industrial & Mfg. Sys. Engg., Ames, IA, United States of America, kazemzad@iastate.edu, Sarah Ryan Unit commitment seeks the most cost effective generator commitment decisions to meet net load while satisfying operational constraints. Stochastic programming and robust optimization are the most widely studied approaches under uncertainty in the load less variable generation. We investigate and compare the performance of these approaches for a multi-bus power system in different aspects including economic efficiency as well as the risk associated with the decisions. 5 - Real-time Energy Disaggregation using Online Convex Optimization Yu Zhang, University of Minnesota, 1033 29th Ave SE, Apt B, Minneapolis, MN, 55414, United States of America, zhan1220@umn.edu, Georgios Giannakis By decomposing a whole electricity consumption into appliance-level signals, energy disaggregation can induce end users’ saving behavior and significantly improve energy efficiency. Capitalizing on underlying features of the sparse and low-rank signals, an online convex optimization problem is formulated for the real-time disaggregation task. An efficient online algorithm is developed with provably sublinear regret. Numerical results corroborate the merits of the proposed approach. Software Demonstration Cluster: Software Demonstrations Invited Session 1 - LINDO Systems, Inc. - Optimization Modeling Made Easy Mark Wiley, VP Marketing, LINDO Systems, Inc. Come and learn how easy it is to: • Quickly build linear, nonlinear, quadratic, conic and integer optimization models, • Incorporate uncertainty into optimization models, • Easily access data from Excel and databases, • Seamlessly embed a solver into your own application. Come and see a demonstration of the power and flexibility of the new releases of: • LINDO API – a callable solver engine, • LINGO – an integrated modeling language and solvers,• What’s Best! – a large-scale solver for Excel. 2 - DO ANALYTICS - OPTEX Mathematical Modeling System: The New Paradigm Jesus Maria Velasquez, Chief Scientist, Do Analytics LLC DO ANALYTICS presents OPTEX Mathematical Modeling System, a powerful expert system that is changing the way to make large scale mathematical programming models. OPTEX: * Generates programming codes in the most powerful optimization technologies, including the SQL statements to connect any DBMS. * Mixes the power of an optimization technology with the easiness of EXCEL. * Works as a client & as an optimization server in the cloud. * Easy and Fast, OPTEX represents the new generation to DO ANALYTICS MD79 79-Room 302, CC

We determine the buyback price for a seller in a purchasing postponement environment. Under stochastic demand a buyer postpones its purchasing decision to reduce inventory cost. The seller on the other, offers a buck back rate to induce higher orders from the buyer. Our model suggests that in a decentralized SC under purchasing postponement, a buy back rate can be arrived at such that the SC profits are maximized and SC coordination is established. 4 - Impact of Supply Relationship Dynamics on Firm Performance: A Multilevel Empirical Analysis Marcus Bellamy, Assistant Professor, Boston University Questrom We develop an empirical model to examine supply relationship dynamics as drivers of firm performance. We use supply chain relationship and financial data from the Bloomberg database. Our unique dataset allows us to investigate manufacturing firms both as customers and suppliers. We use a multilevel mixed- effects model combining firm and dyad level effects. 5 - Logistics Performance Improvement from Information Integration Sung-tae Kim, Assistant Professor, SolBridge International School of Business, 128 Uam-ro, Dong-gu, Daejeon, 300-814, Korea, Republic of, stkim1@solbridge.ac.kr, Gi-eyun Seo This study examines the moderating effects of strategic and operational information integration on the relationships between logistics performance and organizational performance. This study measures logistics performance, in terms of effectiveness, efficiency, and differentiation. Organizational performances are classified as operational, financial, and market performances. The data from 321 manufacturing firms are evaluated using moderated hierarchical regression analysis. Chair: Yu Zhang, University of Minnesota, 1033 29th Ave SE, Apt B, Minneapolis, MN, 55414, United States of America, zhan1220@umn.edu 1 - Predicting and Mitigating Congestion for an Electric Power System under Uncertainty Dzung Phan, IBM T.J. Watson Research Center, 1101 Kitchawan Road, P.O. Box 218, Yorktown Heights, NY, 10598, United States of America, phandu@us.ibm.com, Soumyadip Ghosh Operation of a transmission grid has to handle increasing renewables uncertainty. This necessitates probablistic modeling of the impact of uncertainty over the near- future state of the grid. We propose a multi-period optimization model to estimate the probability of the occurrence of a transmission line congestion event. The model also helps to choose the best mitigation decisions to minimize the chances of experiencing a congestion. A distributed algorithm is presented to efficiently solve it. 2 - Optimal Operation and Services Scheduling for AA Electric Vehicle Battery Swapping Station Hrvoje Pandzic, Faculty of Electrical Engineering and Computing University of Zagreb, Unska 3, Zagreb, Croatia, Hrvoje.Pandzic@fer.hr, Mushfiqur Sarker, Miguel Ortega-vazquez For a successful rollout of electric vehicles (EVs), it is required to establish an adequate charging infrastructure. Battery swapping stations are poised as effective means of eliminating the long waiting times associated with charging the EV batteries. These stations are mediators between the power system and their customers. This presentation describes an optimization framework for the operating model of battery swapping stations. School of Business, 595 Commonwealth Avenue, Boston, MA, 02215, United States of America, bellamym@bu.edu, Soumen Ghosh, Manpreet Hora MD78 78-Room 301, CC Optimization under Uncertainty with Energy Applications Contributed Session

257