Mikael Rnnqvist

1
Integrated routing and transportation planning

• Mikael Rönnqvist
• Division of Optimization, Linköping University,
  Linköping, Sweden
• The Forestry Research Institute of Sweden,
  Uppsala, Sweden

2
Outline

• Forest supply chain and transportation planning
• Problems and models
• Experiences from analyses
• Åkarweb an operational system
• Flowopt strategic / tactical tool
• Concluding remarks

3
Forest supply chain and Transportation planning
4
Forest supply chain
5
Supply chain planning modules
6
Some facts

• The forest industry represents a major part (25)
  of the demand for transportation of goods in
  Sweden.
• The cost of transportation represent one third of
  the total cost of raw material, round wood, to
  the forest industry. It requires a substantial
  use of fossil fuels.
• An efficient use of transportation is vital to
  the competitiveness and environmental influence
  of the forest industry.

7
Characteristics

• Many different transport organisations
• Forest companies, hauliers, individual trucks
• Decentralised planning
• Route decisions are often made by truck drivers
• Many levels of decisions before route planning
• Destinations of logs, allocation of transport
  orders, weekly demand at industry, route decision

8
Destination, transport order
Estimated volumes of assortments
9
Transportation planning

• Destination of logs
• Which assortment to which industry?
• Decisions made by forest company once harvesting
  starts at harvest area.
• Allocation of transport orders
• Which haulier is responsible for transportation
  from entire harvest area to each of the
  industries?
• Decision made by transport manager at forest
  company or responsible haulier.
• Route planning
• Construct daily or weekly routes.
• Descision made by each truck driver (often) or
  haulier responsible for a number of trucks.

10
Models
11
Classical transportation problem
12
Scheduling problem daily routing of trucks
13
Mathematical model scheduling problem
14
Backhaulage tours
Hallstavik
Heby
15
Backhaulage planning
loaded
empty
16
Mathematical model backhaulage problem
17
Experiences
18
Backhaulage planning

• Results
• Cost - 6
• Empty driving - 30
• Time - 9
• Fuel - 9

19
Case study Transportsamordning Nord

• Results
• backhaulage, 46 of volumes
• Empty driving - 24
• Time 9
• Fuel 7

20
Case Södra Skog Backhauling
No sawmills
With sawmills
21
Åkarweb
22
Holmen Skog

• 4.5 million cubic meter harvested volume (2.5 in
  own forest)
• Large proportion of volumes in trading with other
  companies
• Transportation carried out by a number of
  independent companies and organisations
• Development of web based system, Åkarweb,
  started in 1999

23
Åkarweb - purpose

• Decentralised information to transport planners
• Correct information given quickly online
• Co-ordination and integration of several
  transporters
• Co-ordination and integration between companies
• Provide potential backhaulage trips

24
Åkarweb current usage

• 50 transporters associated with Holmen Skog
• 10 transporters associated with Norra Skogsägarna
• Total of 180 trucks connected
• Daily updating (supply/harvest reporting)
• Daily backhaulage planning
• Increased proportion of backhaulage tours (Frisk
  (2003) 15 less empty driving).
• Estimated saving is 6-7 of transportation cost

25
Assortments
Volumes
Industry/ destination
Potential backhaulage
Harvest areas
26
Transport order
Potential backhaulage trips connected to a supply
node
Saved distance
27
(No Transcript)
28
FlowOpt
29
System development - FlowOpt

• Skogforsk Linköping University
• Vinnova
• Holmen Skog, SCA, StoraEnso, Södra
• Optimal Solutions Dianthus
• Project running 2002-2004

30
Route selection with NVDB

• Length
• Road class
• Speed limit
• Surface
• Road width
• Owner

31
Integrated supply chains
32
Model development

• Trains and railway terminals
• Many more variables constraints
• Several participating companies
• Special restrictions on e.g. capacity, trading
• Several planning periods
• Control of storage, thawing
• Aim to use general modeling languages
• Aim to use standard optimization modules

33
Maps
34
Catchment area
35
Testcase

• 322 harvest areas
• 1125 piles of assortments
• 57 industries
• 87 demands

36
Concluding remarks
37
Concluding remarks

• Large potential for savings with centralised
  planning
• Large interest for optimization based planning
  systems
• Åkarweb
• shown to be a practical tool for transport
  planning. Increased potential of backhaulage
  trips
• FlowOpt
• Strategic and tactical (and operational) tool
  truck and train integration, terminal location,
  destination, backhauling planning.
• Practical use through GIS, NVDB and quick result
  generation
• Other systems under development
• RuttOpt Scheduling of trucks
• VägRust Road maintenace / upgrading