Informs Annual Meeting 2017

MA65

INFORMS Houston – 2017

MA63

can be explored with clinical data understand host-microbiome interactions in human health. Subspace estimation forms the basis for many tools for discovering important patterns. Using the traditional sum-of-squared error criterion can lead to erroneous conclusions. We demonstrate the performance of a robust subspace fitting procedure in ordination of microbiome samples and as a subroutine in a subspace classification method. 2 - Dynamic Management of Healthcare Queues with Capacity Adjustment Delay Uncertainty Sara Nourazari, California State University, Long Beach, 1250 Bellflower Blvd-MS.4904, HHS2-118, Long Beach, CA, 90840, United States, sara.nourazari@csulb.edu, Rifat Sipahi, James Benneyan In spite of the incentives to improve outpatient access, many healthcare systems are still struggling with providing timely access to care, which takes its toll on the patients in need. This work provides a detailed analysis pertaining to Modified Smith Predictor, to develop a robust controller design in queue management problems to regulate queue length despite uncertain demand and delay mismatches between the model and the actual system. By illustrating how flexibility in capacity adjustment can effectively improve patient appointment access, the results of this study demonstrate the value of dynamic modelling in healthcare systems redesign initiatives. 3 - The Efficiency Loss Problem in Social Network Analysis Based on Stochastic Block Model Peiwen Xu, City University of Hong Kong, 88 Tat Chee Avenue, Hong Kong, Hong Kong, peiwenxu-c@my.cityu.edu.hk, Tsui Kwok Leung In this paper, we will investigate the efficiency loss for anomaly detection due to the conversion of count data to binary data in social network analysis, based on Stochastic Block Model(SBM), a benchmark for detecting changes of community structure in social network analysis. By applying statistical process control techniques to Stochastic Block Model, the loss of efficiency can be quantified by comparing the average run length (ARL) performance. 4 - Simulation-based Optimization of Prognostics and Health Management enabled Condition-Based Maintenance Policy Omoleye Taiwo Joel, City University of Hong Kong, Kowloon, Hong Kong, jtomoleye3-c@my.cityu.edu.hk, Kwok Leung Tsui The practical implementation of condition-based maintenance (CBM) is trailing behind the widespread theoretical benefits in the literature. This necessitates the growing concern about the effectiveness of CBM policy in industries implementing CBM. To overcome this, research is focusing on the value of Prognostics and Health Management (PHM) information in CBM policy. In this paper, simulation-based optimization is used to assess the benefits of PHM- enabled CBM policy under resource dependence. Also, a framework is developed, numerically illustrated, and compared with classic CBM, and traditional maintenance policies using ARENA-based discrete event simulation model. 370F Design of Logistics Networks Sponsored: TSL, Freight Transportation & Logistics Sponsored Session Chair: Sasha Zhijie Dong, Texas State University, San Marcos, TX, United States, sasha.dong@txstate.edu 1 - A Lagrangian Relaxation Approach to a Multi-Echelon Consolidation of Perishable Products Christine Nguyen, Northern Illinois University, DeKalb, IL, 60115, United States, cnguyen@niu.edu Our research focuses on a supply chain of suppliers with low demand for perishable products. By shipping through FTL units, the transportation cost would be less expensive than LTL shipments. The suppliers all ship to a shared consolidation center, where suppliers can experience the advantages of consolidating product. Transportation and inventory decisions are made at suppliers and the consolidation center. This problem is decomposed and solved using a Lagrangian Relaxation approach. 2 - Regional Logistics Network Design with Consider of Roadway Traffic Congestion and Service Decay Mi Gan, Southwest Jiaotong University, Chengdu, China, taosiyu@swjtu.edu.cn, Si Chen, Siyu Tao In order to solve the problem that the existed logistics network design models(LND) are lack of consideration on roadway traffic flow. The uncover degree function of logistics facility nodes based on impedance function was built. Then, integrated logistics network design models and corresponding algorithms were proposed on the basis of uncovering degree function. The comparison of general LND models and models we developed by actual case reveals the impact of roadway traffic flow on LND. MA65

370D Systems Modeling Approaches to Energy Analysis Invited: Energy and Climate Invited Session Chair: Benjamin D. Leibowicz, University of Texas-Austin, Austin, TX, 78712-1591, United States, bleibowicz@utexas.edu 1 - The Cost of Policy Uncertainty in Electric Sector Capacity Planning: Implications for Instrument Choice Benjamin D. Leibowicz, University of Texas-Austin, ETC 5.128D, 204 E. Dean Keeton St. C2200, Austin, TX, 78712-1591, United States, bleibowicz@utexas.edu The OSeMOSYS energy system optimization model is reformulated as a stochastic program and applied to electric sector capacity planning under climate policy uncertainty in the ERCOT area. Optimal hedging strategies are identified, and the costs of policy uncertainty under carbon taxes, carbon caps, and renewable portfolio standards are compared. Findings clearly indicate that uncertainty considerations favor price-based over quantity-based policy instruments. 2 - Optimal Pathways to Net-zero Emissions: A Study of Austin, Texas Max Brozynski, University of Texas-Austin, Austin, TX, United States, brozynski@utexas.edu We develop an integrated systems model of energy supply and demand at the urban scale, and use it to determine optimal decarbonization pathways. Our framework is integrated in that it represents all major energy end-use demands, and captures synergistic linkages among them. Examples of such linkages include electrification of transportation, vehicle-to-grid technologies, combined heat and power, and diverse uses of hydrogen. We apply our model to evaluate decarbonization strategies in Austin, Texas, which recently adopted a goal to achieve net-zero greenhouse gas emissions by 2050. 3 - Improving Solar Integration and Grid Utilization via Residential Cooling Thermal Energy Storage Thomas Deetjen, University of Texas-Austin, Austin, TX, United States, tdeetjen@gmail.com This study examines the potential for cooling thermal energy storage (CTES) in the residential sector to provide flexibility to the electric grid. A linear optimization model chooses the capacity and sub-hourly dispatch of chillers and CTES systems in a neighborhood with rooftop solar. The model tests whether CTES can reduce peak electric demand, lessen solar-induced ramp rates, and lower operation costs in the neighborhood and bulk transmission-grid. Results show that CTES operation can improve or worsen these outcomes, depending on how electricity rates have been structured. 4 - Optimal Dispatch and Equipment Sizing of a Community-scale Water Recycling Facility for Electric Demand Flexibility Scott Vitter, University of Texas-Austin, scott.vitter@gmail.com, Thomas Deetjen, Bruk Berhanu, Michael Webber This study explored tradeoffs between sizing, operating schedule, and electric load-shifting for a community-scale water recycling facility within a residential area. A mixed-integer linear program was formulated to represent a community system consisting of a batch recycling process, potable water storage, and sewer equalizing storage. Initial results indicate that operating a water recycling facility can reduce overall community costs under certain economic conditions, and that batch processes can be scheduled in response to signals that favor electric demand flexibility. Total water demand, water and sewer pricing, and equipment capital costs strongly influence the optimal solution. 370E New Advancements in Using Data Analytics for Healthcare Applications Sponsored: Data Mining Sponsored Session Chair: Talayeh Razzaghi, Department of Industrial Engineering, New Mexico State, Department of Industrial Engineering, New Mexico S, Las Cruces, NM, 88003, United States, talayehr@nmsu.edu Co-Chair: Lisha Yu, City University of HongKong, City University of HongKong, Kowloon, Hong Kong, lishayu2-c@my.cityu.edu.hk 1 - L1-norm Subspace Estimation for Microbiome Data Analysis Paul Brooks, Virginia Commonwealth Univ, Dept of Stat Sci and OR, P.O. Box 843083, Richmond, VA, 23284, United States, jpbrooks@vcu.edu DNA from communities of microorganisms can be sequenced to provide a census of the species present. The resulting counts of DNA fragments form profiles that MA64

155