Bart JOURQUIN and Sabine LIMBOURG

1
GRT conference, May 7, 2007
Optimal location of container terminalsThe case
of a hub system in Europe

• Bart JOURQUIN and Sabine LIMBOURG
• Catholic University of Mons (FUCAM)
• Group Transport Mobility
• Mons Belgium
• gtm_at_fucam.ac.be

2
Introduction European transport sector and
policy Methodology Application Conclusions -
Prospects
European transport sector and policy

• Major problems
• congestion
• environmental nuisance
• accidents.
• Objectives
• restoring the balance between modes of transport
  and developing intermodality
• Marco Polos objective Decrease of 12.109 t.km
  by road per year
• combating congestion
• putting safety and the quality of services at the
  heart of our efforts
• maintaining the right to mobility.

3
Introduction Combined rail-road
transport Methodology Application Conclusions -
Prospects
Combined rail-road transport
Consolidate flows
Source UIRR
Terminals location crucial
4
Introduction Methology Terminal
typology Application Conclusions - Prospects
Terminal typology

• Ballis (2002)
• Wiegmans (2003)
• Bontekoning and E. Kreutzberger (2001)
• Wiegmans, Masurel and Peter Nijkamp (1998)
• Daubresse (1997)
• SIMET (1995)

T
T
5
Introduction Methology Hub-and-spoke
network Application Conclusions - Prospects
Hub-and-spoke network

• 3 constraints
• all the hubs are connected directly to each
  other
• no direct connection between non-hub nodes
• spoke nodes are connected to a single hub.
• Problem classP-hub Median Problem (P-HMP)
• OKelly (1987)
• Campbell (1994)
• Ernst and Krishnamoorthy (1996)
• Potential location
• Arnold (2002)
• Macharis (2004)
• New feature Systematic approach based on
  transport flows

6
Optimal terminal locations
Introduction Methology A four steps
methodology Application Conclusions - Prospects
Reference assignmentWaterways Roads Railways
0 Data
Supply
Demand
1 Identification
Potential locations
2 P-HMP
Intermodal in an Hub-and-spoke network
3 H-S impact
Final assignment Waterways Roads Railways
Intermodal
7
Introduction Methology Application
Demand Conclusions - Prospects

• Freight OD matrixes for the year 2000 provided by
  NEA
• Roads, railways and inland waterways
• NST-R chapter 9 (diverse commodities)
• Region-to-region at NUTS 2 level
• Most European countries.

8
Introduction Methology Application
Demand Conclusions - Prospects
9
Introduction Methology Application
Supply Conclusions - Prospects
Supply DCW based network with associated
transport costs
10
Introduction Methology Application Calibrated
reference scenario Conclusions - Prospects
Virtual networks
U1 (W2)
U2 (W1)
U3 (R1)
Terminal
U1 (W2 1350T))
U1 (W1 300T)
U3 (R1 Train)
11
Introduction Methology Application Calibrated
reference scenario Conclusions - Prospects
Virtual networks
T
12
Introduction Methology Application Calibrated
reference scenario Conclusions - Prospects
Virtual networks
D
O
Generation
Distribution
Modal split
Virtual Network
Assignment
13
Introduction Methology Application Calibrated
reference scenario Conclusions - Prospects
Behaviour No Yes Capacity
No All or Nothing Stochastic Yes Equilib
rium Stochastic equilibrium
14
Introduction Methology Application Calibrated
reference scenario Conclusions - Prospects
Aggregated demand data No Yes C
apacity No All or Nothing Multi-Flow Yes
Equilibrium Equilibrium MF
15
Introduction Methology Application Calibrated
reference scenario Conclusions - Prospects
Multi-modal, multi-flows assignment
16
Introduction Methology Application
Consolidation Conclusions - Prospects
Consolidated flows on road networks
17
Introduction Methology Application Set of
potential locations Conclusions - Prospects
Set of potential locations

• Possible criteria
• Minimum flow threshold
• Maximum distance to railways
• Minimum distance to existing terminal
• Minimum distance to port
• Maximum distance to waterways.

18
Introduction Methology Application Set of
potential locations Conclusions - Prospects
Set of potential locations
19
Introduction Methology Application
Hypotheses Conclusions - Prospects

• Collection and synthesis
• Real Cost Reduction of Door-to-door Intermodal
  Transport (2001)
• Prospects of Inland Navigation within the
  enlarged Europe (2004)
• Comité National Routier français
• Ministère de la Mobilité des Pays-Bas (2005)

Hypotheses

• (1) Transhipment cost 3.29 /ton
• (2) Inter-hub discount 10
• (3) Pre- and post-haulage 1.483 x long haul
  road cost

(3)
(3)
(1)
(1)
(2)
Source UIRR
20
Introduction Methology Application Inter-hub
networks Conclusions - Prospects
Inter-hub networks
3 terminals
4 terminals
2 terminals
5 terminals
6 terminals
7 terminals
21
Introduction Methology Application Existing
situation in 2002 Conclusions - Prospects
Existing situation
Existing situation -1,34.109 t.km by
road Marco Polos objective -12.109 t.km by
road
22
Introduction Methology Application P-HMP
Optimal locations Conclusions - Prospects
Optimal location
Optimal location -7,59.109 t.km by
road Existing situation -1,34.109 t.km by
road Marco Polos objective -12.109 t.km by road
23
Introduction Methology Application Conclusions -
Prospects

• Major contributions
• Flow based approach
• Methodology for potential locations
• Decision support tools embedded in a GIS.

24
Introduction Methology Application Conclusions -
Prospects
Prospects

• Sensitivity analysis
• Trimodal terminals
• Short-sea shipping