A twoechelon location model: the WEEE recovery system

1
A two-echelon location model the WEEE recovery
system

• Maria Isabel Gomes Salema FCT-UNL, Portugal
• Ana Paula Barbosa Póvoa IST, Portugal
• Augusto Novais INETI, Portugal

2
Outline

• Problem context and definition
• The model
• Several analysis
• Network designs
• The location of new recycling facilities
• Final remarks

3
Problem context

• Waste of electrical and electronic equipment
  (WEEE)
• Directive 2002/96/EC enforces producers to
  support collecting and recycling costs of their
  products

• 10 categories of products
• Large household appliances
• Small household appliances
• IT and telecommunications equipment
• Consumer equipment
• Lighting equipment
• Electrical and electronic tools
• Toys, leisure and sports equipment
• Medical devices
• Monitoring instruments
• Automatic dispensers

4
Problem context - Amb3e

• Portuguese Association for the recovery of WEEE
• Created by 57 EEE producers (2004)
• Core function
• Organize and manage the WEEE recovery network
  (collection, transportation, treatment and
  recycling)

5
Problem definition

• Design the WEEE recovery network, assuring legal
  targets imposed by the EU
• Entities
• WEEE sources
• Sorting centres
• Recyclers
• Products
• Centres inbounds are unsorted WEEE
• Five different flows of WEEE are sent to
  recyclers

6
The model

• Location-allocation model for the reverse supply
  chain sorting-centre locations
• Capacitated model
• Multi-product
• Multi-period time is modelled using a two-unit
  structure
• Transportation modes

7
Time

• Modelled through a management perspective
• Horizon
• Strategic decisions facility location
• Horizon is divided in macro units
• Tactical decisions return satisfaction
• Macro time is divided in smaller units (micro
  time)
• Tactical decisions collection, storage and
  recycling planning

8
Problem formulation

• Variables

Flow variables amount of product m supplied by
entity i to entity j at micro time t These model
flows between entities, disposal flows and
recycling amounts
continuous
Stock variables amount of product m stored in
facility i at micro time t
Non-satisfied collection amount of product m not
collected at source i at macro time t
9
Problem formulation
10
Case data

• 278 municipalities integrate the network
  representing
• WEEE sources
• possible locations for sorting centres
• Amb3e set the collection target to 3
  kg/inhabitant per year

11
Performed analysis

• Year 2006
• Uncapacitated model
• Recycling capacity limited
• Time horizon of five years (assuming an growth of
  3 per year in the WEEE volume)
• Uncapacitated model
• Recycling capacity limited
• Location of new recycling facilities

12
Uncapacitated network (Year 2006)

• 73 (out of 278) municipalities are chosen to
  locate sorting centres

13
Uncapacitated network (Year 2006)

• 73 (out of 278) municipalities are chosen to
  locate sorting centres
• WEEE sources
• 268 are entirely satisfied
• There is no collection in the remaining 10,
  which represent only 0.9 of the total WEEE
  generated
• The location of the major recycling unit in
  Tondela has an enormous impact on the network
  structure

14
Recycling planning

• There is almost no lighting equipment to recycle
• The total amount sent to disposal equals the
  maximum volume allowed by legislation

15
Computational results
GAMS/CPLEX (built 22.0) Pentium IV 3.4 GHz
16
New locations for recyclers

• Given the current facility operating in Tondela,
  4 new locations are considered
• 3 near Porto (north)
• 1 in Alentejo (south)
• Goal
• Select the new locations for recycling facilities
• Select the sorting centres location
• Define the recycling plan in all facilities (new
  and existing)
• Define the collection and storage plans

17
New locations for recyclers

• Recycling facilities
• 2 (out of 4) are selected
• Tondela is still operating
• Sorting centres
• 70 (out of 278) municipalities are chosen
• WEEE sources
• 274 are entirely satisfied
• Only 4 have no collection

18
New locations for recyclers
19
Computational results
GAMS/CPLEX (built 22.0) Pentium IV 3.4 GHz
20
Final remarks

• Study of a Portuguese recovery network for WEEE
• Locations of sorting centres and new recycling
  facilities
• Flow, recycling, freight planning
• Sensitivity analysis on some parameters (e.g.
  penalization cost, disposal cost)
• Design of a new network level, between sorting
  centres and recyclers

21
Missing setence...

• The European Union (EU) estimates a growth of 3
  to 5 per year for the electrical and electronic
  equipment waste (WEEE),

22
A two-echelon location model the WEEE recovery
system

• Maria Isabel Gomes Salema FCT-UNL, Portugal
• Ana Paula Barbosa Póvoa IST, Portugal
• Augusto Novais INETI, Portugal