Session Detail Information
Add this session to your itinerary

Joint Clusters :  Transportation Science & Logistics
 Railroad Applications

Session Information  : Sunday Oct 24, 10:00 - 11:30

Title:  Joint Session TSL/Railroad App.: Mathematical Programming-Based Marine Transportation and Logistics
Co-Chair: Marco Duran,ExxonMobil Research & Engineering, 1545 Route 22 East, Annandale NJ 08801, United States, marco.a.duran@exxonmobil.com
Chair: Kevin Furman,ExxonMobil Research & Engineering, 1545 Route 22 East, Annandale NJ 08801, United States, kevin.c.furman@exxonmobil.com

Abstract Details

Title: Integrated Inventory Management and Ship Scheduling of LNG
 Presenting Author: Marielle Christiansen,Professor, Norwegian University of Science and Technology, Alfred Getz vei 1, Trondheim N-7491, Norway, Marielle.Christiansen@iot.ntnu.no
 Co-Author: Ole-Tobias Frich,Norwegian University of Science and Technology, Alfred Getz vei 1, Trondheim N-7491, Norway, frich@stud.iot.ntnu.no
 Richard Horgen,Norwegian University of Science and Technology, Alfred Getz vei 1, Trondheim N-7491, Norway, richarho@stud.iot.ntnu.no
 
Abstract: The gas field Snøhvit (Snow White) north of Norway will start its production next year. We have developed a tactical planning model for Statoil where we consider both the production and storages close to Snøhvit and the transportation of LNG (liquefied natural gas) to terminals in Europe and America. The underlying mathematical model is an integrated inventory management and ship scheduling model, where spot trade is included in the model. It is solved by use of standard optimization software.
  
Title: Robust Models for Tank Container Repositioning
 Presenting Author: Martin Savelsbergh,Georgia Institute of Technology, School of Ind and Sys Engineering, Georgia Institute of Technology, Atlanta GA 30332, United States, mwps@isye.gatech.edu
 Co-Author: Alan Erera,Georgia Institute of Technology, School of Ind and Sys Engineering, Atlanta GA 30332-0205, United States, alerera@isye.gatech.edu
 Juan Morales,Georgia Institute of Technology, School of Ind and Sys Engineering, Atlanta GA 30332-0205, United States, jmorales@isye.gatech.edu
 
Abstract: Tank containers are a safe and cost-effective way to transport liquid chemical products globally. To cost-effectively serve customer demands which are by nature geographically and temporally imbalanced, tank fleet operators utilize asset repositioning models. We develop new weekly repositioning models that determine robust plans given uncertain customer requests but do not rely on probabilistic information.
  
Title: Marine Inventory Routing
 Presenting Author: David Ronen,Professor, University of Missouri-St. Louis, College of Business Administration, One University Place, St. Louis MO 63121, United States, david.ronen@umsl.edu
 
Abstract: Marine inventory routing problems are faced by oil and chemical companies and other manufacturers of large quantities of liquid bulk products. This presentation will: (1) characterize the marine inventory routing problem, (2) Differentiate it from the familiar inventory routing problem, (3) provide classification of marine inventory routing problems, (4) review the limited research done in this area, and (5) identify areas where additional research has large potential leverage.
  
Title: Optimization and Logistics of Combined Marine Transportation/Inventory Systems
 Presenting Author: Marco Duran,ExxonMobil Research & Engineering, 1545 Route 22 East, Annandale NJ 08801, United States, marco.a.duran@exxonmobil.com
 Co-Author: Lifei Cheng,Carnegie Mellon University, Chemical Engineering, Pittsburgh PA, United States, lifei_cheng@merck.com
 
Abstract: We present a decision support framework, based on the integration of discrete event simulation and stochastic optimal control, for the optimization and logistics of combined marine transportation/inventory systems. We apply the approach to a complex and large-scale world-wide crude oil marine transportation problem. We formulate the problem as a Markov decision process. Near-optimal control policies are obtained using approximate methods based on Dynamic Programming principles.