Informs 2015

171 / 552

INFORMS Philadelphia – 2015
169
4 - A Network Design Model for Dual-channel Commerce with
Uncertain Demand
Guoqing Zhang, Professor, University of WIndsor,
401 Sunset, Windsor, ON, Canada,
We study the supply network problem with dual-channel, where a manufacturer
or a retailer sells their products from both direct channel and traditional retail
channel. An optimization model is established to examine central and local
distribution/fulfilling center strategies, and determine inventory levels at each
location and market allocation with taking into account uncertain demands from
both channels.
MA70
70-Room 202A, CC
Rail Safety and Risk Analysis
Sponsor: Railway Applications
Sponsored Session
Chair: Xiang Liu, Assistant Professor, Rutgers University, CoRE 606,
96 Frelinghuysen Road, Piscataway, NJ, 08854-8018,
United States of America,
1 - Using Text Mining and Data Visualization to Analyze Railroad
Grade Crossing Accidents
Trefor Williams, Professor of Civil Engineering, Rutgers
University, 96 Frelinghuysen Rd, Piscataway, NJ, 08540,
United States of America, , John Betak
This paper will discuss how probabilistic topic modeling, clustering, text
visualization and data visualization have been used to study the nature of
accidents that occur at grade crossing. The data analyzed were from the Federal
Railroad Administration grade crossing accident database. The paper will illustrate
how trucks, particularly tractor-trailer trucks have been identified as a major
cause of highway-railroad accidents from analysis of the text fields of grade
crossing accident reports.
2 - Comparative Train Accident Analysis for Class I
U.S. Freight Railroads
Zhao Wang, Graduate Research Assistant, University of Illinois at
Urbana-Champaign, 205 N. Mathews Ave., B-118 Newmark
Engineering Laboratory, MC, Urbana, IL, 61801, United States of
America, Christopher Barkan,
Mohd Rapik Saat
This paper studies the U.S. Class I freight train accident rates between 2004 and
2013. By utilizing train accident data and statistical theories, the investigation
compares and contrasts train accident rates, trends, accident cause distributions,
and accident severity between two time periods. The decreasing trend in accident
rate and the changing trend in accident cause distributions are analyzed.
3 - Fault Tree Analysis of Train Accidents on Shared-use Rail Corridor
Chen-Yu Lin, Graduate Research Assistant, University of Illinois
at Urbana-Champaign, 205 N. Mathews Ave., Urbana, IL, 61801,
United States of America, Mohd Rapik Saat,
Christopher Barkan
Safety issues regarding shared-use rail corridors are emerging due to the
introduction of faster and more frequent passenger trains. In this research, a
general risk framework is presented for shared-use rail corridors. A standard risk
management procedure is implemented to identify potential hazards and evaluate
the risk associated with them. Fault tree analysis is performed to hazards with
higher level of risk. An example fault tree for a specific hazard, adjacent track
accident, is presented.
4 - Risk-based Rail Inspection and Repair
Xiang Liu, Assistant Professor, Rutgers University, CoRE 606,
96 Frelinghuysen Road, Piscataway, NJ, 08854-8018,
United States of America,
Broken rails are the leading causes of freight-train derailments in the United
States. Each year, the railroad industry spends millions of dollars on rail
inspection and repair. This research develops a simulation-based risk analysis
model to optimize ultrasonic rail inspection and repair strategies.
MA71
71-Room 202B, CC
Alternative Fuel Vehicles and Sustainable
Transportation II
Sponsor: TSL/Urban Transportation
Sponsored Session
Chair: Jee Eun Kang, Assistant Professor, University at Buffalo,
409 Bell Hall, Buffalo, NY, United States of America,
1 - Locating Battery Exchange Facilities on Lines and Trees
Pitu Mirchandani, Arizona State University, Tucson, AZ,
United States of America, Yazhu Song
We introduce a new location problem for battery exchange facilities (BEF) for
Electric Vehicles. First we study the problem of location feasibility on lines. Given
feasibility, the location problem becomes “where should BEFs be located to
minimize a charge-related objective”. e.g., the objective of minimizing the
maximum distance between BEFs minimizes the anxiety of the drivers. Scenarios
include single OD pair, multiple OD pairs, round trips, etc. Extensions to tree
networks is discussed.
2 - Design for EV Market Systems
Namwoo Kang, Research Fellow, University of Michigan, 2350
Hayward Street, Ann Arbor, MI, 48109, United States of America,
Panos Papalambros, Fred Feinberg
Electric Vehicle (EV) market systems are generally run by three key players: EV
manufactures, charging station operators, and government. This study presents a
decision-making framework for the players to maximize profits and greenhouse
gas reductions by quantitative modeling and linking of consumer demand
prediction (marketing), charging station sitting (operations), EV powertrain
design (engineering), and investment allocations (public policy). Several scenarios
and case studies are examined.
3 - Optimal Deployment of Charging Lanes in
Transportation Networks
Zhibin Chen, University of Florida, 365 Weil Hall, Gainesville,
United States of America, , Fang He,
Yafeng Yin
This paper attempts to develop a mathematical model to optimally deploy in a
large-scale highway network charging lanes that charge electric vehicles while
they are on the move. We first describe network flow equilibrium conditions
under a particular deployment plan of charging lanes, and then formulate the
design of charging lanes as a mathematical program with equilibrium constraints.
4 - Modeling Intra-household Interactions for the Use of Battery
Electric Vehicles
Yashar Khayati, Graduate Research Assistant, State University of
New York at Buffalo, 327 Bell Hall, Buffalo, NY, 14226, United
States of America, , Jee Eun Kang
This study assesses the potential use of Battery Electric Vehicles in place of
conventional Internal Combustion Engine Vehicles at household level. A
sequential activity allocation and insertion heuristic is developed to implement on
HAPPEV. The results show that if BEVs would be used at household level the
travel disutility of households can be decreased about $42 per day in average. In
comparison, if a BEV is used to do exact same activity pattern the average saving
for the day is only $7.
MA72
72-Room 203A, CC
2015 QSR Best Student Paper Competition
Sponsor: Quality, Statistics and Reliability
Sponsored Session
Chair: Eunshin Byon, Assistant Professor, University of Michigan,
1205 Beal Avenue, Ann Arbor, MI, 48109, United States of America,
1 - 2015 QSR Best Student Paper Award
Eunshin Byon, Assistant Professor, University of Michigan, 1205
Beal Avenue, Ann Arbor, MI, 48109, United States of America,
Best Student Paper Award recognizes excellence among QSR student members.
Four finalists for the Best Student Paper Award will make presentations. The
winner will be announced at the QSR business meeting during the conference.
MA72