Informs 2015

95 / 552

INFORMS Philadelphia – 2015
93
4 - Spotting Anomalies in Cyber Physical Datasets:
The Case of Mobility Data
Konstantinos Pelechrinis, Assistant Professor, University of
Pittsburgh, 135 N. Bellefield, IS 717B, Pittsburgh, PA, United
States of America, , Evangelos Papalexakis,
Christos Faloutsos
How can we discover latent patterns in heterogeneous CPSs datasets and classify
them as anomalous or not without labeled data? We propose using tensors to
model heterogenous data and obtain latent patterns. We then propose a generic
data-driven method for classifying each of the obtained patterns as normal or not.
The realization of our technique is domain-specific. We showcase our method by
applying it on a mobility dataset that captures locations visited by users at
different times.
SC03
03-Room 303, Marriott
Contemporary Scheduling
Cluster: Scheduling and Project Management
Invited Session, Chair: Joseph Y.T. Leung
Distinguished Professor, New Jersey Institute of Technology,
Department of Computer Science, University Heights, Newark, NJ,
07102, United States of America
1 - Improved Algorithms for Single Machine Scheduling with Release
Dates and Rejections
Kangbok Lee, York College, CUNY, York College, The City
University of New York, Jamaica, NY, 11451, United States of
America, , Cheng He, Joseph Leung,
Michael Pinedo
We consider bi-criteria scheduling problems on a single machine with release
dates and rejections and both the makespan and the total rejection cost have to be
minimized. We consider two scenarios: (i) minimize the sum of the makespan and
the total rejection cost, and (ii) minimize the makespan subject to a bound on the
total rejection cost. We summarize the results obtained in the literature and
provide for several cases improved approximation algorithms and FPTASs.
2 - Integrated Production and Delivery on Parallel Batching Machines
Kai Li, Associate Professor, Hefei University of Technology, 193
Tunxi Rd, Hefei, 230009, China, Joseph Leung,
Zhao-hong Jia
We consider an integrated scheduling problem of production and delivery on
parallel patching machines. The company will earn a positive profit only if a job is
delivered by its due date. A 3PL provider is used to deliver the jobs. The goal is to
maximize the total profit. We show that the problem is solvable in polynomial
time if the jobs have identical sizes, but it becomes unary NP-hard if the jobs have
different sizes. We propose heuristics for NP-hard cases and analyze their
performances.
3 - Minimizing Total Completion Time in Flow Shop with Machine
Unavailability using Meta-heuristics
Hairong Zhao, Associate Professor, Purdue University at Calumet,
Dept. of Math, C. S, & Statistics, Hammond, IN, 46323, United
States of America, , Yumei Huo
We consider flow shop scheduling subject to machine availability constraints. The
objective is to find a schedule that minimizes total completion time. This problem
is strongly NP-hard even if machines are always available. Simple bounds are
derived to slightly speed up the elimination process of a branch-and-bound
algorithm. Then we propose a meta-heuristic algorithm based on genetic
algorithms. Computational results show that the proposed meta-heuristic
performs effectively and efficiently.
4 - Application of MGSA for the Coordinated Scheduling Problem in a
Two-Stage Supply Chain
Jun Pei, Assistant Professor, Hefei University of Technology, 193
Tunxi Rd, Hefei, 230009, China, ,
Xinbao Liu
This paper investigates a products and vehicles scheduling problem in a two-stage
supply chain, where jobs need to be processed on the serial batching machines of
multiple manufacturers distributed in various geographic zones and then
transported by vehicles to a customer. A modified gravitational search algorithm
(MGSA) is proposed to solve the problem. In MGSA,Several improvement
strategies and the batching mechanism DP-H are introduced.
SC04
04-Room 304, Marriott
Gender Inspired Research
Sponsor: Women in OR/MS
Sponsored Session
Chair: Margret Bjarnadottir, Assistant Professor of Management Science
and Statistics, Robert H. Smith School of Business, University of
Maryland, 4324 Van Munching Hall, College Park, MD, 20742,
United States of America,
1 - Innovative Pedagogical Interventions to Increase Retention of
Women in Engineering
Susana Lai-yuen, Associate Professor, University of South Florida,
4202 East Fowler Avenue, ENB 118, Tampa, FL, 33620,
United States of America, , Grisselle Centeno
This work addresses the broad challenge of identifying practices and developing
resources to help overcome evident gender equity issues in science and
engineering education. Specifically, a set of pedagogical resources focused on
healthcare systems and OR applications have been developed. Experiences related
to development, implementation and outcomes will be discussed.
2 - Bridging the Gap: Responses to Equal Pay Legislation
David Anderson, Assistant Professor, Baruch,
Margret Bjarnadottir
We study how firms can reduce the estimated pay gap between men and women
in the most cost efficient way. We show that by intelligently increasing workers’
wages who will have the greatest impact, we can meet the “Equal Pay for Equal
Work” standard for less than half the cost of the naive method of increasing all
female workers’ wages equally. We further explore the impacts of equal cost
mandates on compensation, fairness and the implications of this work on outside
verification parties.
3 - Work-Life Balance for Women
Wendy Casper, The University of Texas at Arlington, Dept. of
Management, Arlington, TX, United States of America,
, Victoria Chen
Work-life balance is important to many women. Despite this, there is little
agreement about what work-life balance is. This presentation discusses the notion
of balance and identifies commonly held definitions of this concept, concluding
with a few ideas about how women can gain a greater sense of balance in their
lives.
SC05
05-Room 305, Marriott
Extracting Business Value from Social Media
Analytics: Techniques and Applications
Cluster: Social Media Analytics
Invited Session
Chair: Shih-Hui Hsiao, University of Kentucky, 550 S. Limestone,
Lexington, KY, 40526, United States of America,
1 - Who are the Opinion Leaders? A Relative Assessment of Opinion
Leader Mining Algorithms
Shih-Hui Hsiao, University of Kentucky, 550 S. Limestone,
Lexington, KY, 40526, United States of America,
Ram Pakath
Several methods have been proposed in the Social Media Analytics literature for
identifying Opinion Leaders (OL) in online social networks. In this talk, I will
describe the design and implementation of, as well as preliminary findings from,
an experiment that compares existing OL mining algorithms to one another in
terms of solution speed and quality. This study is a prelude to a larger project that
also seeks to improve upon extant procedures.
2 - Real-time Social Media Analytics in Health Care:
Discovery Knowledge from Online Communities
Yichuan Wang, Industrial & Systems Engr St, Department of
Industrial and Systems Engineering, Shelby Center,
Auburn University, Auburn, AL, 36849 ,
Yedurag Babu, Terry Byrd
Dramatic changes in business environments have galvanized firms toward
searching for external knowledge from social media to complement the
insufficiency of organizational resource. However, in healthcare social media
sources rarely have been analyzed and used to support medical decision making.
This study proposes a real-time knowledge discovery framework to support
effective exploration of knowledge which has been prototypically implemented
on the base of Web data from healthcare communities.
SC05