Informs 2015

255 / 552

INFORMS Philadelphia – 2015
253
MD66
66-Room 113C, CC
Aviation Applications Section: Keynote Presentation
Sponsor: Aviation Applications
Sponsored Session
Chair: Senay Solak, University of Massachusetts Amherst, Isenberg
School of Management, Amherst, MA, 01003, United States of
America,
1 - FAA’s Modernized Terminal Area Forecast
Dipasis Bhadra, Economist, FAA, 800 Independence Avenue, SW,
Washington, DC, 20591, United States of America,
The Terminal Area Forecast tool is designed to integrate views of local, national
and international flow of activities and capture effects of socioeconomic and
technological factors on aviation. The tool is used to understand airports,
passenger routing, and aircraft network impact of NEXTGEN development. It also
provides projections for future air transport activity through time using future
passengers by origin and destination (O&D) market routes and networks (i.e.,
segment flows); aircraft operations by markets and network routes; and integrates
operations and passenger flows through the National Airspace System (NAS)
network. The forecasting tool is used to help understand the policies, procedures,
and environmental regulations.
MD67
67-Room 201A, CC
Integrated Vehicle Routing Problems II
Sponsor: TSL/Freight Transportation & Logistics
Sponsored Session
Chair: Weihong Hu, Georgia Tech, Atlanta, GA,
United States of America,
1 - Heuristics for an Integrated Inventory Routing and Freight
Consolidation Problem of Perishable Goods
Weihong Hu, Georgia Tech, Atlanta, GA, United States of
America, Alejandro Toriello,
Maged Dessouky
We study a novel mixed integer programming model that integrates freight
consolidation and inventory routing of perishable goods. We propose an iterative
framework that combines a decomposition procedure involving three
subproblems and an optimization-based local search scheme. Experiments based
on empirical distributions of real data demonstrate the effectiveness of both
solution approaches for small to medium size problem instances. We further
extend the approach for larger problem instances.
2 - Reoptimization Capabilities of Benders Decomposition for the
Stochastic Production Routing Problem
Yossiri Adulyasak, HEC, Canada,
Jean-Francois Cordeau, Raf Jans
We present two approaches that exploit the reoptimization capabilties to speed up
the Benders decomposition algorithms for the the production routing problem
(PRP), which is a generalization of the inventory routing problem (IRP), under
demand uncertainty. The first approach is applied to the two-stage stochastic PRP
in the context of a sample average approximation (SAA) method. The second
approach is embedded into a rollout algorithm for the multi-period stochastic PRP.
Computational results are presented.
3 - Resource-Constrained Dynamic Programming with “Hot-Starting”
for the Elementary Shortest
Ahmad Jarrah, George Washington University, Washington, DC,
United States of America, Luis Novoa,
Xinhui Zhang, Jonathan Bard
We develop the complete theory for designing a dynamic program (DP) for
solving elementary shortest path problems with idle time cost. This is integrated
with bidirectional extensions, decremental state-space relaxation, 2-cycle
elimination and sharpest-to-date restricted sets of unreachable nodes. We describe
new hot-starting procedures to significantly improve the DP’s run time. The
approach can be used in solution algorithms for the capitated VRPTW problem
with idle time cost.
MD68
68-Room 201B, CC
Joint Session TSL/Public Sector: Resilience in
Transportation Infrastructure Systems
Sponsor: Transportation, Science and Logistics
Sponsored Session
Chair: Lili Du, Assistant Professor, Illinois Institute of Technology, 3201
S. Dearborn Street, Chicago, IL, United States of America,
1 - A System-of-Systems Approach toward a Resilient, Dynamically
Interdependent Transportation Network
Elise Miller-Hooks, Professor, University of Maryland, College
Park, MD, , Neza Vodopivec
A transportation network’s vulnerability to damage depends on the vulnerabilities
of other critical systems (i.e. energy, telecommunications, building infrastructure).
In a disaster, interdependencies both within a system and between different
systems are reshaped dynamically as people take adaptive actions to mitigate
impact and repair networks. We explore the interactions between changes in
network structures and the evolution of interdependencies between systems.
2 - Infrastructure Investment Decisions in Multimodal Intercity
Transportation Networks
Bo Zou, University of Illinois at Chicago, 2095 Engineering
Research Facility, 842 W. Taylor Street (M/C 246), Chicago, IL,
60607-7023, United States of America, , Lili Du,
Mohamadhossein Noruzoliaee
Infrastructure investment in multimodal intercity transportation networks
involves maintaining existing capacity and adding new capacity to infrastructure
facilities. To achieve maximum returns, the infrastructure planner should
understand the behavior of players in the network, especially their responses to
investment. We develop a bi-level framework to model the decision process,
where the upper level pursues social welfare maximization subject to lower-level
supply-demand equilibrium.
3 - Optimal Location and Operation of Railroad First-response
Resources
Siyang Xie, PhD Student, University of Illinois at Urbana-
Champaign, Urbana, IL, United States of America,
, Yanfeng Ouyang
Railroad incidents such as derailments often seriously impact a large region and
block traffic in nearby highway networks, which in turn hinders efficiency of first
response efforts. In this paper, we develop a reliable optimization model to
characterize and guide positioning and utilization of railroad first-responder
resources. Customized solution techniques are employed to effectively solve the
model and to provide decision support.
4 - Hazmat Network Design using Time-dependent Consecutive
Road Closures Considering Intermediate Stops
Tolou Esfandeh, University at Buffalo, SUNY, Department of
Industrial & Systems Eng., 339A Bell Hall, Buffalo, NY, 14260,
United States of America, ,
Changhyun Kwon, Rajan Batta
we analyze the regulator’s problem of identifying the sections of the network and
their corresponding times that should be closed to hazmat transportation in a
time varying network. We assume that the closure of a section is consecutive in
time and the hazmat carriers are allowed to stop in the middle of their trip. We
develop a column generation algorithm that accounts for routing and scheduling
alternatives which not only reduce the risk but also accounts for drivers’ cost
perspectives.
MD68