2015 Informs Annual Meeting

WE55

INFORMS Philadelphia – 2015

WE55 55-Room 108B, CC Design Engineering Contributed Session Chair: Jayash Koshal, Bloomberg LP, 425 Washington Blvd, Jersey City, NJ, 07310, United States of America, jayashkoshal@gmail.com 1 - Performance Improvement of Average Access Time and Cache Memory Access Time Mohammad Shamsu Alam, Assistant Engineer, ADN Telecom, 17, Red Crescent Concord, Tower, Mohakh, 121/1&2, Tejkoni Para, Tejgaon, Dhaka, 1215, Bangladesh, shiblyshamsul@gmail.com, Mirza Moazzem Hossain, Jamal Uddin Ahmed The cache is a small amount of high speed memory. Cache is placed between the CPU and main memory. In this view, we consider in this paper to optimize the cache memory access time which eventually optimizing the average access time and also if the cache memory size doubled then the cache memory access time should also double but cache memory access time will only increase 10% from the previous value. 2 - The Thermal Contact Resistance Testing Method Study of Thin Film Materials Xiaogang Li, BUAA, Rm 318, Weimin Building, 37 Xueyuan Rd., Haidian Dist., Beijing, China, lxg@buaa.edu.cn As there are many factors affecting the thermal contact resistance, In engineering application, the thermal contact resistance of the solid interfaces is usually obtained by testing. In this paper, the method of measuring the thermal contact resistance of the thin film materials is presented. According to this method, we use copper heat flux meter, graphite columnar specimens and graphite thin film sample to measure the thermal contact resistance between the graphite thin film material 3 - Data-enhanced Prediction of Marketing Strategies: Does Product Repairability Play a Role? Mostafa Sabbaghi, Graduate Research Assistant, University at Buffalo, SUNY, 437 Bell Hall, Buffalo, NY, 14260-2000, United States of America, mostafas@buffalo.edu, Sara Behdad The existing manufacturers’ policies in making products difficult and expensive to repair and not adequately share the repairing information with end-users and independent repair businesses make repair a challenging process. This paper aims to characterize the product sales rate and post purchase usage behavior to investigate the impact of product reparability on consumers’ future purchase decisions and the long-term profitability of manufactures. 4 - Response Surface Modeling Approaches for Multi-response Design Parameter Optimization Problems Gölten Gügayaz, Research Assistant, Middle East Technical University, Industrial Engineering Department, Ankara, Turkey, ggulten@metu.edu.tr, Gulser Koksal Performances of response surface modeling approaches such as Ordinary Least Squares, Weighted Least Squares, Generalized Least Squares, Multivariate Regression, and Seemingly Unrelated Regression are analyzed for multiple- response design parameter optimization problems. Weaknesses are discussed and improvements are suggested. 5 - Maximum Entropy Utility Function Assessment using Renyi Entropy Jayash Koshal, Bloomberg LP, 425 Washington Blvd, Jersey City, NJ, 07310, United States of America, jayashkoshal@gmail.com, Vijitashwa Pandey Maximum entropy methods, used to assess probability distributions, can also be used for assessing utility. When utility values are known for discrete values of the attributes, a problem arises of fitting a curve to these points. We use Renyi entropy instead of Shannon’s entropy to allow different weighting of different regions of utility functions. We show how Maximum Renyi Entropy formulation can be used for utility functions and compare the results with Maximum Shannon Entropy functions.

WE56 56-Room 109A, CC Manufacturing III Contributed Session

Chair: Hossein Badri, PhD Student, Wayne State University, Industrial and Systems Engineering Dep, Detroit, MI, 48201, United States of America, hossein.badri@wayne.edu 1 - A Spatiotemporal Outlier Detection Framework for Diagnostic Quality Monitoring of Image Sequences

Mohammad Abed Motasemi, UTSA, 6685 UTSA Blvd., San Antonio, TX, 78249, United States of America, abed.motasemi@gmail.com, Adel Alaeddini

Statistical quality monitoring has mainly remained with less attention in images and high dimensional data application. We propose a novel spatiotemporal outlier detection framework for monitoring the quality of image sequences. Extensive analysis based on a real application in a manufacturing facility is conducted to demonstrate the performance of our proposed methodology in detecting various types of outliers. 2 - The Cause and Effect of Material Handling Systems on Operational Performance Yunjung Suh, PhD Candidate, Kyungpook National University, 102-1001 Maechunro 2Gil, 19, Bukgu, Daegu, 702-793, Korea, Republic of, yunjsuh@knu.ac.kr, Jaewoo Chung This study investigates the causal relationships among actual firm’s motives, degrees of material handling automation systems(MHS) and operational performance based on surveys in Korea. The study classifies and measures the usage of MHS and identifies actual firms’ motives to introduce MHS with a fit to the firm’s business strategy. The result presents that there is an associative relationship among firm’s strategy, MHS and operational performance with the mediating effect of MHS. 3 - Shifting Bottleneck Heuristic for Stochastic Job Shop Scheduling Problems Hossein Badri, PhD Student, Wayne State University, Industrial and Systems Engineering Dep, Detroit, MI, 48201, United States of America, hossein.badri@wayne.edu, Milad Zafarnezhad, Kai Yang Job shop scheduling problems are among the hardest known combinational optimization problems. In this research a shifting bottleneck heuristic algorithm is developed for stochastic job shop scheduling problems. To evaluate the performance of the developed algorithm, the heuristic is tested with randomly generated instances. The results indicate that the developed heuristics performs very well for minimizing makespan in stochastic job shop problems. Chair: Zhi Zhou, Argonne National Laboratory, 9700 South Cass Ave. Bldg 221, Lemont, Il, 60437, United States of America, zzhou@anl.gov 1 - Two-stage Robust SCUC with Mixed-integer Recourse Lei Wu, Clarkson University, 8 Clarkson Ave, Potsdam, United States of America, lwu@clarkson.edu This talk presents a two-stage robust SCUC approach to adaptively and securely respond to continuous load/wind uncertainty intervals and discrete N-K contingency security criteria. Corrective capabilities of both non-quick-start and quick-start units are formulated. The proposed model is solved by the combination of modified BD method and C&CG algorithm. Case studies illustrate the effectiveness of the proposed robust SCUC approach. 2 - Flexible Operation of Batteries in Power System Scheduling with Renewable Energy Audun Botterud, Argonne National Laboratory, 9700 S Cass Ave, Lemont, IL, United States of America, abotterud@anl.gov, Nan Li, Emil Constantinescu, Canan Uckun, John Birge, Kory Hedman We present a two-step framework to evaluate the value of energy storage in power systems with renewable generation. First, we formulate a stochastic unit commitment model with wind power forecast uncertainty and energy storage. Second, we derive a flexible schedule for energy storage in economic dispatch with limited look-ahead horizon. A case study demonstrates the benefits of battery storage in systems with renewable resources and the effectiveness of the proposed operational strategy. WE57 57-Room 109B, CC Stochastic Methods Applied to Power System Operations Sponsor: ENRE – Energy I – Electricity Sponsored Session

496