2016 INFORMS Annual Meeting Program

WE01

INFORMS Nashville – 2016

Wednesday, 4:30PM - 6:00PM

optimizing crew assignment. Numerical experiment demonstrates that the proposed algorithm can increase the flight stability, meanwhile minimize the total disruption cost induced. 2 - Developing A Dynamic Tool For Transplant Survival Analysis Hamidreza Ahady Dolatsara, PhD Candidate, Auburn University,

WE01 101A-MCC Data Mining in Manufacturing Sponsored: Data Mining Sponsored Session Chair: Mojtaba Khanzadeh, Mississippi State University, 21 Ace Avenue, 21 Apartments, Starkville, MS, 39759, United States, mk1349@msstate.edu 1 - A Congestion Prediction-based Dynamic Routing Model In Automated Material Handling Systems Sang Min Lee, Senior Researcher, Samsung Electronics, 816 ChanguiKwan, Korea Univ., 145 Anam-ro, Seongbuk-gu, Seoul, Korea, Republic of, smlee5679@gmail.com, Jee Hyuk Park In automated material handling systems of semiconductor manufacturing, vehicular congestion is a persistent problem resulting in the reduction of production efficiency. In order to effectively route vehicles to reduce traffic congestion, this study presents a congestion-avoidance routing model based on congestion prediction. A congestion prediction model is proposed to predict the possibility of probable heavy congestion that can lead to significant production loss. The effectiveness of the proposed model is demonstrated by using real data supplied by a semiconductor fabrication plant in South Korea. 2 - Profile Monitoring And Fault Diagnosis Via Sensor Fusion For Multi-stream Data Weihong Guo, Assistant Professor, Rutgers, The State University of New Jersey, 96 Frelinghuysen Rd, CoRE Rm 220, Piscataway, NJ, 08854, United States, wg152@rutgers.edu When multiple signals are acquired from different sources, sensor fusion and data dimension reduction are two major issues to achieve a better comprehension of the process. Methods for analyzing multi-stream profiles based on multilinear discriminant analysis and ensemble learning are proposed in this research for the purpose of profile monitoring, fault detection, and fault diagnosis. The proposed methods are compared with state-of-the-art methods with both simulated and real data. WE02 101B-MCC Data Mining Applications Sponsored: Data Mining Sponsored Session Chair: Leily Farrokhvar, West Virginia University, 395 Evansdale Drive, Morgantown, WV, 26506, United States, leily@vt.edu 1 - An Analysis Of Charitable Givings And Donor Behavior Negar Darabi, Graduate Student Researcher, West Virginia University, Morgantown, WV, 26506, United States, nedarabi@mix.wvu.edu, Leily Farrokhvar, Azadeh Ansari While charitable givings are typically a noticeable portion of the GDP and there is abundant data available through tax forms, there has been few systematic studies to identify contributing factors and predict donor behavior. Additionally, disasters are shown to have a significant temporary effect on charitable givings. In this study, we analyze the historic data using regression models to identify the most influential factors and analyze impact of natural disasters on donor behavior.

Suite 3301, 345 West Magnolia Ave., Auburn, AL, 36849, United States, hamid@auburn.edu, Ali Dag, Bin Weng, Fadel Mounir Megahed

This study present a toll developed for three types of survival analysis for the transplants. In the first type, it estimates if a patient could survive certain time windows which are integer multiples of one year. As the second type, it yield probability of survival. This tool also estimates expected survival time. Surgeons or other practitioners could utilize it based on their available data from their patients and donors. These data are collected during the registration, waiting list, operation, and after the operation. The tool utilizes machines learning methods to identify the importation features and then utilizes the features for model training and delivering a requested analysis. 3 - An Optimization Approach To Detection Of Epistatic Effects Maryam Nikouei Mehr, PhD Student, Iowa State University, 3004 Black Engineering, Ames, IA, 50011, United States, mnmehr@iastate.edu, Lizhi Wang Epistasis refers to the phenomenon where the interaction of multiple genes affects a certain phenotype more than they do separately. Similar epistatic effects are also ubiquitous in other application areas, where a certain effect is only observable when a particular combination of multiple factors is present. Due to the enormous solution space, it’s hard to detect the epistatic effect. We propose an optimization model that attempts to detect epistatic effects where a large number of observations are available for a relatively small number of explanatory factors. We will share our preliminary results and discuss future research directions. 4 - Warehouse Process Improvement Through Data Analytics And Optimization Vedat Bayram, Postdoctoral Research Fellow, University of Waterloo, 200 University Ave. West, Waterloo, ON, N2L 3G1, Canada, vbayram@uwaterloo.ca, Fatma Gzara, Samir Elhedhli With the advance of the technologies on collecting and storing data, warehouses, from the most automated to the manual, generate large amounts of data. Warehouse management companies are seeking ways to get the full value from the massive amounts of data and use it as a competitive advantage in the marketplace. In this presentation, we report on data analytics solutions for a warehouse management and control systems company. We develop descriptive tools to analyze the big data of e-retailing warehouses and identify process improvement opportunities. We develop data driven optimization solutions and validate by comparing to real system operation. WE04 101D-MCC Capacity-Expansion Planning with Increasing Renewable Levels Sponsored: Energy, Natural Res & the Environment, Energy I Electricity Sponsored Session Chair: Ramteen Sioshansi, Ohio State University, 1971 Neil Avenue, Columbus, OH, 43210, United States, sioshansi.1@osu.edu Co-Chair: Antonio J. Conejo, The Ohio State University, 1971 Neil Avenue, Columbus, OH, 43210, United States, conejonavarro.1@osu.edu 1 - Analyzing European Climate And Energy Policy Using Stochastic Optimization Asgeir Tomasgard, NTNU, asgeir.tomasgard@iot.ntnu.no The paper presents a modeling based analyses of decarbonization options for the European power sector. Different support schemes designed to incentive early development of CCS are studied, like public grants, feed-in premiums and emission portfolio standards are evaluated. As an alternative we study storage and transmission expansion in combination with a high renewable share. For the analysis we use the EMPIRE model, a multi-horizon stochastic investment model for the European power system that combines long-term capacity expansion with operational modeling under different load and generation scenarios.

WE03 101C-MCC Big Data I Contributed Session 1 - Crew Assignment Subject To Flight Delay Risks

Hing Kai Chan, Associate Professor, University of Nottingham Ningbo China, 199 Taikang East Road, Room AB260, Ningbo, 315100, China, hingkai.chan@nottingham.edu.cn, Sai Ho Chung, Jing Dai Crew cost ranks as the second highest cost of flight operations, but failure of assigning sufficient crew members to a flight will lead to flight disruption such as delay. The dilemma is obvious. This study adopts a big-data approach by utilizing historical flight arrival delay data and a learning algorithm to predict such risks for

484