4 présentations

• 
  10h30 - 10h55

  Consolidation based model for the Scheduled Service Network Design Problem

  • Fabien Lehuédé, prés., IMT Atlantique, LS2N
  • Mike Hewitt, Loyola University Chicago

  The Scheduled Service Network Design Problem is solved by transportation companies to establish a tactical planning of vehicle moves in a freight transportation network. It involves planning the transportation of multiple commodities from terminals to terminals in the network, minimizing costs through consolidation at hubs and taking account of travel time as well as the commodity release dates and due dates. This problem is classically formulated and solved on a time-expanded network, involving advanced decomposition methods to address problems of realistic sizes. In this presentation, we introduce a new modeling approach which avoids creating a time-expanded network by considering all feasible consolidations on the network arcs. We show that this new formulation outperforms the classical model when solved by a solver. We also show that it is more appropriate to model some asset management consideration, bin backing constraints and piecewise linear costs that are often realistic in this context. Finally, we present several reinforcement and hybridation of the consolidation based formulation that make its solving even more efficient in term of run time or addressed problem size.
• 
  10h55 - 11h20

  Proactive shipment consolidation: order allocation and planning based on the trade-off between purchasing costs and transportation costs

  • Bara Najat, EMINES-Mohammed VI Polytechnic University
  • Ozturk Onur, prés., Telfer School of Management, University of Ottawa

  Shipment consolidation is the process of grouping different shipments in a larger one to gain economies of scales. This work is inspired from a real company that deals with the electric works in large construction projects (malls, train stations…). It procures most of the material from suppliers abroad. Depending on the agreed Incoterm, either the selected supplier deals with transportation or one of the company carriers does. The company does local optimizations to select the cheapest offer (the supplier and the Incoterm) for each order. This policy leads to the dispersion of orders in space over several regions and in time over spaced delivery dates which decreases the transportation efficiency. This paper presents a proactive strategy for shipment consolidation on time and space. We have developed a mixed integer linear model which selects the suppliers based on the trade-off between the purchasing costs, transportation costs and the storage costs. The model distinguishes between already launched orders for which only the truck is determined, and new orders to be launched for which the region (supplier), the picking time (week) and the truck must be determined.
• 
  11h20 - 11h45

  Stochastic Scheduled Service Network Design: the Value of Flexible Schedules

  • Mike Hewitt, prés., Loyola University Chicago
  • Simon Belieres, HEC Montréal

  Scheduled service network design (SSND) problems are key issues in tactical transport planning. However, while transport systems are inherently subject to many sources of uncertainty, most of the SSND-related literature focuses on deterministic variants. We investigate the value of modelling flexibility offered by transportation modes such as trucking. We focus on a case where the carrier commits to the physical network before knowing shipment volumes. Recourse decisions determine (i) the timing of the operations while matching the physical paths and point-to-point services set in the first stage as well as (ii) the potential use of extra ad-hoc capacity. We introduce a formulation of the Stochastic Scheduled Service Network Design with Flexible Schedules (SSSNDFS) and present a custom solution method for its solution. Through computational experiments, we observe that modelling flexible schedules generate significant savings, particularly for large instances. A qualitative comparison reveals management strategies for leveraging flexibility. We also observe that modelling flexible schedules result in structurally different solutions in terms of physical network design. Finally, we illustrate the effectiveness of the proposed method.
• 
  11h45 - 12h10

  Integrating Tactical and Operational Planning in Scheduled Service Network Design

  • Saeedeh Dehghani, prés., CIRRELT
  • Teodor Gabriel Crainic, Université du Québec à Montréal
  • Walter Rei, Université du Québec à Montréal

  This study proposes a top-down approach to integrating tactical planning decisions into operational planning for the scheduled service network design in freight transportation. Our primary objective is to optimize revenue generation while ensuring customer demand and requirements are met, and resources are efficiently utilized. We consider the interactions between the carrier's services and operations to enhance the value chain of freight transportation systems and foster stakeholder collaboration. Our study addresses the challenges of transitioning from tactical to operational planning, which is relatively unexplored in the field of freight transportation.

  This research will contribute to the field of scheduled service network design by proposing a comprehensive framework that integrates tactical and operational planning while promoting collaboration and negotiation among stakeholders. The results of our study will provide a proof of concept for the proposed framework and highlight its potential to improve the performance of freight transportation systems through enhanced decision-making and resource utilization. By bridging the gap between tactical planning and operational planning, our research offers valuable insights for practitioners and researchers in the field of intermodal freight transportation.