2016 INFORMS Annual Meeting Program

SB67

INFORMS Nashville – 2016

SB64 Cumberland 6- Omni Applications and Methodological Issues on MCDM Sponsored: Multiple Criteria Decision Making Sponsored Session Chair: Danielle Morais, Universidade Federal de Pernambuco, Recife - PE, Brazil, daniellemorais@yahoo.com.br Co-Chair: Adiel Teixeira De Almeida Filho, Universidade Federal de Pernambuco, Management Engineering Department, Recife, Brazil, adieltaf@cdsid.org.br 1 - A Navy Weapon Selection Throughout Fitradeoff Adiel Teixeira De Almeida Filho, Assistant Professor, Universidade Federal de Pernambuco, Recife, Brazil, adieltaf@cdsid.org.br, Leonardo A Pessoa, Rodrigo Ferreira, Adiel Teixeira de Almeida This work presents a multiple criteria decision model for selecting a weapon to be incorporated in a navy ship using the FITradeoff method. A numerical application is presented based on realistic data with regard to the real problem faced by a Brazilian Navy. 2 - A Multicriteria Model For Supplier Selection Based On A Multilinear Utility Function Felipe Macedo de Morais Pinto, Universidade Federal de Pernambuco, felipe_mmp94@hotmail.com, Adiel Teixeira De Almeida Filho This work presents an MCDM model based on a multilinear utility function for selecting a maintenance service supplier. Depending on the context, maintenance activities may need to consider other criteria besides cost, which are detailed in the reference Multicriteria and Multiobjective Models for Risk, Reliability and Maintenance Decision Analysis. 3 - Computing Interval Weights For Incomplete Pairwise Comparison Matrices Of Large Dimension – A Weak Consistency Based Approach Jana Krejcí, PhD Student, University of Trento, Via Sommarive 9, Povo, Trento, 38123, Italy, jana.krejci@unitn.it Jana Krejcí, PhD Student, University of Bayreuth, Universitat sstr. 30, Bayreuth, D-95440, Germany, jana.krejci@unitn.it, Vera Jandova, Jan Stoklasa, Michele Fedrizzi We present a novel interactive algorithm for large-dimensional pairwise- comparison problems based on the sequential optimal choice of the pairwise comparisons (PCs) to be provided by the decision maker and the concept of weak consistency. The proposed solution significantly reduces the number of needed PCs by providing sets of feasible values for all missing PCs after each input of a new PC. Interval weights of objects covering all possible weakly consistent completions of the incomplete PCMs are then computed from the resulting incomplete weakly consistent PCM. The algorithm is capable of reducing the number of PCs required in PC matrices of dimension 15 and greater by more than 60% on average. SB65 Mockingbird 1- Omni Learning Analytics of Massive Open Online Courses (MOOCs) Sponsored: Information Systems Sponsored Session Chair: Sang Pil Han, Arizona State University, Arizona State University, Tempe, AZ, 85281, United States, sangpil78@gmail.com 1 - Cohort Size And User Engagement: A MOOC Field Experiment Jiye Baek, Boston University, jiyebaek@bu.edu Jesse C Shore MOOCs have the potential to transform how people access knowledge, but they face substantial difficulties in keeping users engaged. We conduct a field experiment on the edX platform to identify factors that promote student engagement in MOOC discussion forums, focusing on cohort size. While most prior work show that users in smaller groups participate more per person, our results show that in the MOOC, the students in larger size cohorts interact more per person and that this greater interaction in turn increases student retention and performance. We theorize that larger cohorts produce more forum content and thus increase the resources available to draw marginal students into an engaged state.

2 - Towards Improved Education For Students Of Low Socioeconomic Status: Learning Analytics Of Massive Open Online Courses (MOOCS) Sang Pil Han, Arizona State University, Main Campus, PO Box 874606, Office:BA 301D, Tempe, AZ, 85287-4606, United States, sangpil78@gmail.com, Mi Hyun Lee, Sunghoon Kim, Sungho Park Although the new era of free, online learning unfolds, the claim of ‘education for all’ appears to be overshadowed by the concern over the unequal use of Massive Open Online Courses (MOOCs). MOOCs may not be a viable solution to students across all levels of socioeconomic status (SES). Using learner outcome and demographic data at a MOOC, we examine the effectiveness of two intervention strategies to improve engagements among low SES learners: (1) course verification which allows learners to earn an official credit later and (2) mobile media which enable learners to attend MOOCs anytime/anywhere. From the findings, we draw implications that can help expand access to education to everyone through MOOCs. SB66 Mockingbird 2- Omni QSR Student Introduction and Interaction and Best Student Poster Competition Sponsored: Quality, Statistics and Reliability Sponsored Session Chair: Nan Chen, National University of Singapore, 21 Lower Kent Ridge Road, Singapore, 119077, Singapore, isecn@nus.edu.sg Co-Chair: Kaibo Wang, Tsinghua University, Department of Industrial Eignieering, Tsinghua University, Beijing, 100084, China, kbwang@tsinghua.edu.cn 1 - Student Introduction And Interaction And Best Student Poster Competition Nan Chen, National University of Singapore, isecn@nus.edu.sg This session provides a platform for the interactions between students and senior QSR members. Participating students will present their research in poster and oral presentation form. The best poster will be voted and selected among all posters. We also invite faculty members and industry representatives to interact with students. They will share valuable experience and provide career advice. Chair: Chiwoo Park, chiwoo.park@eng.fsu.edu Co-Chair: Shiyu Zhou, shiyuzhou@wisc.edu 1 - Structural Damage Growth Prediction Via Integration Of Finite Element Method And Bayesian Estimation Approaches Yuhang Liu, Graduate Student Research Assistant, University of Wisconsin–Madison, 1513 University Ave., Madison, WI, United States, liu427@wisc.edu, Shiyu Zhou Damage diagnosis and prognosis play an important role in ensuring the safety of mechanical and civil structures. Existing works are limited to estimation of the damage magnitude at the current time instance. Revealing the evolving path of structural damage is highly desirable for prognosis and remaining useful life prediction. In this paper, we propose a dynamic data-driven hierarchical Bayesian degradation model, which takes advantage of both the physical finite element model and the data driven Bayesian framework, to tackle the structural damage growth prediction. The damage growth trend can be efficient and accurately estimated by Gibbs sampling. Numerical and case studies are presented. 2 - Dynamic Data Driven Visual Surveillance Via Cooperative Unmanned Aerial/ground Vehicles Sara Minaeian, University of Arizona, Systems and Industrial Engineering, Tucson, AZ, 85721, United States, minaeian@email.arizona.edu, Jian Liu, Young-Jun Son Unmanned vehicles (UVs) with onboard sensors have recently shown promising performance in various applications such as autonomous surveillance, compared to the fixed sensors. However due to the uncertain and dynamically changing environment, the complex problem of autonomous crowd control requires robust, multi-scale and effective algorithms to be applied in real-time. In this work, we propose an autonomous visual surveillance system based on dynamic data-driven adaptive multi-scale simulation (DDDAMS) for crowd control in a border area. The experimental results reveal effectiveness of the proposed system in accomplishing assigned missions under dynamic conditions. SB67 Mockingbird 3- Omni Dynamic Data Driven Application Systems Sponsored: Quality, Statistics and Reliability Sponsored Session

65