4 Presentations

• 
  10:30 AM - 10:52 AM

  A GRASP with Variable Neighborhood Descent for the Pickup and Delivery Problem with Time Windows

  • Rodrigue Tchapnga Takoudjou, presenter, University of Bordeaux
  • Jean-Christophe Deschamps, University of Bordeaux
  • Rémy Dupas, Université de Bordeaux, IMS, CNRS 5218

  Transportation is a dominating function in the logistic networks. Real link in the supply chain, the carrier has become a key player in the logistics whose performance strongly influences the performance of suppliers and customers. The pickup and delivery problem with time window (PDPTW) is a special class of transportation problems. The objective of the PDPTW is to satisfy at minimum cost a set of transport requests while respecting a set of constraints. This paper introduces a new Greedy Adaptive Search Procedure (GRASP) which embeds VND (Variable Neighborhood descent) for solving the PDPTW. Our approach is validated by using a set of benchmark instances from the literature.

• 
  10:52 AM - 11:14 AM

  Distribution and Inventory Planning with Uncertain Demand

  • Mikael Rönnqvist, presenter, Norwegian School of Economics and Business Administration
  • P Flisberg, Linköping University – Sweden
  • John Gunnar Carlsson, University of Minnesota

  Distribution and inventory planning is an important planning exercise for many companies. In a general distribution network there may be several transportation modes and nodes representing mills, terminals and customers. Even though there are confirmed customer orders there are often uncertainties on the overall demand of products. The traditional approach to deal with uncertainty is to work with safety stock levels at terminals and/or distribution centers. In this paper, we introduce the use of robust optimization to handle the uncertainty of customer orders and to establish distribution and inventory plans. In the proposed approach, we can use historical information and practical considerations on the uncertainty. A case study from a large pulp producer is used to test the proposed approach against a traditional approach with safety stock. The analysis is based on simulations and it shows that the robust approach is more cost efficient.

• 
  11:14 AM - 11:36 AM

  Flexibility and Consistency Issues in Static and Dynamic Inventory-Routing

  • Leandro C. Coelho, presenter, Université Laval
  • Jean-François Cordeau, HEC Montréal, GERAD, CIRRELT
  • Gilbert Laporte, HEC Montréal

  Inventory-routing problems (IRPs) arise in vendor-managed inventory systems. They require jointly solving a vehicle routing problem and an inventory management problem. We introduce the concept of transshipments as a way to promote flexibility within the IRP. Consistency leads to higher quality of service by regulating the frequency of the deliveries and the quantities delivered. Both concepts are later integrated within a dynamic environment, in which one can take better advantage of them. We propose matheuristics to solve several classes of IRPs with flexibility and consistency features.

• 
  11:36 AM - 11:58 AM

  New Extensions of the Unidirectional Flow Path Design Problem

  • Julie Rubaszewski, presenter, Université de Technologie de Troyes
  • Alice Yalaoui, Université de Technologie de Troyes
  • Lionel Amodeo, Université de Technologie de Troyes
  • Sylvain Fuchs, ANDRA – France

  The flow path design is the determination of each segment direction and the paths to used for automated vehicles in production units. It is necessary to ensure that all pick up and delivery points can be reached from any other points. The organization of these roads directly influences the performance of a system, as the transport time and the number of necessary vehicles. That is why this problem is one of the most important issues in AGV (automated guided vehicles) systems design. This work deals with flow path design problem for a conventional unidirectional network. Some new extensions of the total travel distance minimization considering both loaded and empty travels is presented. This study is made in collaboration with ANDRA (National Agency for Radioactive Waste Management).