Informs 2015

198 / 552

INFORMS Philadelphia – 2015
196
2 - Real-time Airline Schedule Recovery
Dejun Hang, Jeppesen, Englewood, CO,
United States of America,
In airline daily operations, the schedule is often disrupted by events unforeseen at
the planning stage. Recovering the disrupted schedule as fast as possible with
good quality is critical to protect revenue. We present how Jeppesen’s Fleet
Management System approaches this problem. We will discuss the underlying
models and algorithms used to solve the aircraft and crew schedule recovery
problems, and also some of the critical support processes.
MB71
71-Room 202B, CC
Mobility Choices in Urban Transportation
Sponsor: TSL/Urban Transportation
Sponsored Session
Chair: Hai Jiang, Tsinghua University, Dept of Industrial Engineering,
Beijing, China,
1 - Determinants of Private Vehicle use Intensity: Evidence from
Disaggregate Household Data in China
Hai Jiang, Tsinghua University, Dept of Industrial Engineering,
Beijing, China, , Zhao Zhang
The level of congestion is primarily determined by the number of vehicles owned
by the residents and their use intensity (that is, annual vehicle miles driven).
Existing literature typically conduct the analysis using aggregated macroeconomic
variables. In this talk, we use disaggregate household data to identify
determinants of private vehicle use intensity in China. We find that gasoline price
plays little role in use intensity, which is consistent with results from existing
literature using macroeconomic data. We also find that the socioeconomic
characteristics of the household has considerable effect in vehicle use intensity.
2 - A Comparison of Mixed Logit and Latent Class Methods for Mode
Choice Analysis
Yuntao Guo, Purdue University, Lyles School of Civil Engineering,
West Lafayette, IN, 47907-2051, United States of America,
Jian Wang, Srinivas Peeta
This study aims to investigate the differences between the mixed logit and the
latent class methods in terms of model fit, model insights, predicted travel mode
choice, and spatial transferability using a large sample of revealed preference
travel mode choice data. The model can help planners to develop effective
strategies to foster more sustainable transportation mode choice behaviors by
reducing automobile dependency and encouraging the usage of alternative modes
of travel.
3 - Traffic Equilibrium and Pricing with Information in a
Correlated Network
Song Gao, Associate Professor, University of Massachusetts
Amherst, Department of Civil and Env. Engineering, 214C
Marston Hall130 Nat. Resources Rd, Amherst, MA, 01003, United
States of America, , Andre de Palma
We study the impacts of correlation and information penetration rate in a
network with correlated, random link capacities, in terms of the types of
equilibrium solution (corner vs interior for informed and uninformed
respectively), the user cost for informed and uninformed, the social cost, and the
optimal price for system optimum.
MB72
72-Room 203A, CC
Journal of Quality Technology (JQT) Invited Session
Sponsor: Quality, Statistics and Reliability
Sponsored Session
Chair: Fugee Tsung, Prof., HKUST, Clear Water Bay, Kowloon, Hong
Kong - PRC,
1 - An Introduction to Statistical Issues and Methods in Metrology
Joanne Wendelberger, Los Alamos National Laboratory, Statistical
Sciences Group, MS F600, Los Alamos, NM, 87545, United States
of America, , Michael Hamada, Max Morris,
Stephen Vardeman, J. Marcus Jobe, Tom Burr, Huaiqing Wu,
Leslie Moore
Statistical science and metrology provide valuable concepts and approaches for
assessing the quality of measured data. Measurement quality impacts the
knowledge that can be gained by collecting and analyzing data using statistical
methods, and appropriate data collection and analysis quantifies the quality of
measurements. An overview of statistical issues and methods in metrology is
presented that includes both frequentist and Bayesian methodologies.
2 - From Profile to Surface Monitoring: SPC for Cylindrical Surfaces
via Gaussian Processes
Bianca Colosimo, Professor, Politecnico MILANO, via La Masa, 1,
Milano, 20156, Italy,
Massimo Pacella
Quality of machined products is often related to the final shapes of the
manufactured surfaces. This paper presents a novel method for surface
monitoring, which combines Gaussian processes to model the manufactured
shape and multivariate control charting for monitoring the deviations of the
actual surface from the in-control pattern. Regardless of the specific case study,
the proposed approach is general and can be extended to deal with different kinds
of surfaces or profiles.
3 - Multimode Geometric Profile Monitoring with Temporally
Correlated Image Data
Abhishek Shrivastava, Assistant Professor, FAMU-FSU College of
Engineering, Dept of Industrial & Manufacturing Eng,
Tallahassee, FL, 32310, United States of America,
, Park Chiwoo
We propose a new method for monitoring changes in geometrical profiles of
objects in a dynamic process; changes in profiles occur in various modes. This
work is motivated by the need for monitoring changes in geometrical shape and
sizes of nanoparticles during self-assembly process. The proposed multimode
geometric profile monitoring method addresses three specific issues - profiling of
functional data, monitoring of multimode processes, and monitoring of time-
correlated processes.
4 - Statistical Surface Monitoring by Spatial-Structure Modeling
Kaibo Wang, Associate Professor, Tsinghua University,
Department of Industrial Engineering, Beijing, China,
Fugee Tsung, Andi Wang
We propose a new chart based on the Gaussian-Kriging model to monitor a two-
dimensional (2-D) surface. A parametric model that considers three components
of the surface, the global trend, the spatial correlations, and independent errors, is
first constructed; then we monitor the process by detecting changes in the
estimated parameters. This method is utilized to monitor a wafer-manufacturing
process and its performance is compared with that of an existing method through
simulation.
MB73
73-Room 203B, CC
Game-theoretical Models in Maintenance
and Reliability
Sponsor: Quality, Statistics and Reliability
Sponsored Session
Chair: Maryam Hamidi, PhD Candidate, University of Arizona, 1127 E.
James E. Rogers Way, Room 111, P.O. Box 210020, Tucson, AZ, 85721,
United States of America,
1 - A Cooperative Game of Spare-Parts Systems
Ulas Ozen, Ozyegin University, Ozyegin University Cekmekoy
Campus, Istanbul, Turkey,
We consider a group of firms that keep spare-parts inventory to maintain their
equipments. The firms can cooperate by pooling their spare-parts inventory and
reduce costs. One important question is how the benefit of such cooperation
should be shared between the participating firms. The firms’ spare-parts
investment problem is modeled as a queueing model and we study the associated
cooperative game. We show that the resulting game has a non-empty core.
Several extensions are studied further.
2 - Non-cooperative Game Theory Analysis in Supply Chain
Internal Financing
Wanying Shi, Western New England University, 1215 Wilbraham
Street, Springfield, MA, 01119, United States of America,
, Julie Drzymalski
This paper evaluates the effect of wholesale price discount contract on
coordinating non-cooperative supply chain financing between a liquid supplier
and a capital constrained retailer at a risk-free rate. Optimal policies for retailer
and supplier are found. Profits and efficiencies of retailer, supplier and supply
chain are compared in both the retailer- and supplier-led Stackelberg games.
Results show that internal financing will increase overall profits.
MB71