Informs 2015

63 / 552

INFORMS Philadelphia – 2015
61
2 - Airline Trip Segmentation
Manini Madireddy, Sabre, 3150 Sabre Dr, Southlake, TX, 76092,
United States of America, ,
Aditya Kothari, Goda Doreswamy, Sergey Shebalov,
Ramasubramanian Sundararajan
We consider the problem of recommending customized product bundles to airline
customers based on their behavior and preferences. We present an approach to
correlate candidate product designs to traveler behavior, based on historical
evidence of travel and purchase patterns. We use clustering algorithms to segment
travelers and further demonstrate how the goodness of a clustering result can be
measured through the extent to which it enables good product design.
3 - Predicting Operational Performance for Airline Schedules
through Simulation
Mahmood Zangui, Senior Systems Engineer, Optym, 7600 NW
5th Place, Gainesville, FL, 32607, United States of America,
Eric Camacho, Yu-shiu Lin,
Pranav Gupta, Jeff Borges, Lonny Hurwitz
An airline schedule is a highly interconnected network; where one disruption can
propagate through the network and delay several flights. Airlines are interested in
measuring the robustness of their schedules against such disruptions, and in
forecasting their operational performance. We have built a simulation model that
can predict key performance indices for given schedules. We will share our
modeling approach, challenges we faced, and lessons learned during development
of this model.
4 - A Hybrid Model for Airline Flight Frequency and
Scheduling Optimization
Ahmed Abdelghany, Associate Professor, Embry-Riddle
Aeronautical University, 600 S. Clyde Morris blvd, Daytona
Beach, FL, United States of America, ,
Khaled Abdelghany, Farshid Azadian
An optimization model that considers the joint optimization of airline flight
frequency and flight scheduling decisions is presented. The model takes into
consideration the schedule of other competing airlines. Several experiments are
presented to illustrates the model capabilities.
SA67
67-Room 201A, CC
City Logistics and Sustainable Urban
Freight Systems - I
Sponsor: TSL/Freight Transportation & Logistics
Sponsored Session
Chair: Johanna Amaya, Rensselaer Polytechnic Institute, 110 8th St.
JEC 4037, Troy, NY, 12180, United States of America,
1 - Public Sector Initiatives Sustainable Urban Freight Systems Part I
Jeffrey Wojtowicz, RPI, United States of America,
Cara (xiaokun) Wang, Johanna Amaya, Jose Holguin-Veras,
Miguel Jaller, Carlos A. Gonzalez-Calderon, Ivan Sanchez-Diaz,
Stacey Hodge, Michael Browne, Daniel Haake
Transportation policy should ensure freight is moved efficiently as it is a physical
manifestation of the economy. From this study, planners are provided with
guidelines to implement initiatives addressing freight issues. Initiatives were
organized as part of a continuum, from supply to demand side; underpinned by
stakeholder engagement. This paper focuses on supply side and covers
infrastructure management, parking and loading, vehicles, and traffic
management along with stakeholder engagement.
2 - Toward a Freight-friendly Built Environment:
A Mixed Method Approach
Kazuya Kawamura, Associate Professor, University of Illinois at
Chicago, MC348, 412 S. Peoria St., Chicago, IL, 60607,
United States of America, , Takanori Sakai
This study uses both parking violation data for quantitative analysis and video
images collected from pole-mounted portable video recorders for qualitative
contextual assessment to shed light on the relationship between delivery vehicle
activities and the built environment including streetscape in Chicago.
3 - Electric Vehicle Routing Problem
Jane Lin, University of Illinois at Chicago, 842 W. Taylor Street
(M/C 246), Chicago, IL, 60607, United States of America,
, Wei Zhou
This paper presents a general Electric Vehicle Routing Problem (EVRP) that finds
the optimal routing strategy with minimal travel time cost and energy cost as well
as number of EVs dispatched. As demonstrated with a case study in Austin TX,
the effect of vehicle load on routing strategy cannot be ignored. Compared to
diesel truck VRP, EVRP has comparable travel time and distance but long en-route
re-charging time. Lastly, the network topology greatly affects the routing
strategies.
4 - A Plugin Hybrid Electric Vehicle Routing Problem with Recharging
and Refueling
Matthew Roorda, Associate Professor, University of Toronto, Civil
Engineering Department, University of Toronto, Toronto, On,
M5S 1A4, Canada, , Mehdi Nourinejad,
Sina Bahrami, Glareh Amirjamshidi
This paper presents a Hybrid Electric Vehicle Routing Problem which specifically
considers Electric Vehicles and Plugin Hybrid Electric Vehicles. A modified Clarke
Wright algorithm is proposed. The model is applied to the City of Toronto. The
results of the model provide managerial insight. For instance, the analysis shows
the total number of Electric Vehicles is insensitive to the battery capacity but
sensitive to the fuel tank capacity of Plugin Hybrid Electric Vehicles.
SA68
68-Room 201B, CC
Geospatial Analysis in Transportation and Logistics
Sponsor: Transportation, Science and Logistics
Sponsored Session
Chair: Eunsu Lee, Assistant Professor, New Jersey City University, 2039
John F. Kennedy Blvd, Jersey City, NJ, 07305, United States of
America,
1 - Spatial Matching Models: Freight Demand Analyses Considering
Partner Selection and Joint Decision
Dapeng Zhang, Doctoral Student, Rensselaer Polytechnic
Institute, Cara (xiaokun) Wang
Freight demand has not been understood comprehensively due to unique
features: the freight activity is a result of collaboration of multiple agents. In
specific, the collaboration can be observed by partner selection and joint decision
making processes. As existing models cannot capture such features, this research
develops an innovative model, spatial matching model, to fill the void. The
proposed model is specified based on freight behavior, validated by experiments,
and explained by examples.
2 - Stochastic Market Boundaries and Fertilizer Plants
Sumadhur Shakya, Assistant Professor, California State Univ.
Monterey Bay, 100 Campus Center, Seaside, CA, 93955,
United States of America, William Wilson,
Bruce Dahl
The study analyzes spatial competition among nitrogen-based fertilizer plants in
United States, to gain insight on viability of newly announced fertilizer plants and
structure of supply chain, now and in future, at a macro level using locational
information; linear and mixed-integer based optimization techniques and
geographical information systems methods are applied to arrive at current spatial
equilibrium and most likely future cases for structure of supply chain in Untied
States.
3 - An Agent-based Modeling for Remotely Piloted Aircraft (RPAS)
Package Delivery in Rural Area
Ali Rahim Taleqani, Research Assistant, Upper Great Plains
Transportation Institute, Dept. 2880, P.O. Box 6050, Fargo, ND,
58108-6050, United States of America,
, Eunsu Lee
The purpose of this paper is to examine the agent-based modeling concept in the
context of urgently-needed package delivery system by means of Remotely
Piloted Aircraft (RPA). The target market is assumed to be in rural area in Cass
County North Dakota where travelling distance to major markets are longer and
accessibility is not well developed. The implementation is constructed over Agent
Analyst simulation platform in ArcGIS software.
4 - Analysis of Alabama Statewide Freight Flow after the International
Capacity Expansion
Jaehoon Kim, University of Alabama in Huntsville, 301
Sparkman Dr., Technology Hall S239, Huntsville, AL, 35899,
United States of America, Michael Anderson
This study presents a methodology to estimate state level of freight flow. Using
FAF3 database, disaggregated state level O/D matrices are developed and applied
to the State of Alabama. Thereafter, the expected freight flow variation affected by
the Panama Canal expansion is applied to the O/D matrices. The developed O/D
matrices are assigned to the statewide network. The study identifies current
Alabama freight flow pattern and new freight flow trend after the Panama Canal
expansion.
SA68