ng decision-making methodologies that determine
how to maximize the environmental benefits of end-of-
life (EOL) processing while minimizing costs under variable
EOL situations. This paper describes a methodology to
analyze how product designs and situational variables impact
the Pareto set of optimal EOL strategies with the greatest
environmental benefit for a given economic cost or
profit. Since the determination of this Pareto set via
enumeration of all disassembly sequences and EOL fates
is prohibitively time-consuming even for relatively simple
products, multi-objective genetic algorithms (GA) are utilized
to rapidly approximate the Pareto set of optimal EOL trade-
offs between cost and environmentally conscious actions.
Such rapid calculations of the Pareto set are critical to better
understand the influence of situational variables on how
disassembly and recycling decisions change under different
EOL scenarios (e.g., under variable regulatory, infrastructure,
or market situations). To illustrate the methodology, a
case study involving the EOL treatment of a coffee maker
is described. Impacts of situational variables on trade-
offs between recovered energy and cost in Aachen, Germany,
and in Ann Arbor, MI, are elucidated, and a means of
presenting the results in the form of a multi-situational EOL
strategy graph is described. The impact of the European
Union Directive regarding Waste Electric and Electronic
Equipment (WEEE) on EOL trade-offs between energy recovery
and cost was also considered for both locations.
Introduction
Increased demand for consumer electric and electronic
products, combined with the accelerated pace at which
technology is evolving, has inevitably resulted in an increased
amount of obsolete, discarded, broken, or abandoned
products that must be treated by society. Consumer electric
and electronics products are of particular concern due to
high production volumes and characteristically short time
scales of technological or stylistic obsolescence leading to
landfilling of large amounts of discarded product. Exacer-
bating this problem is the fact that the components in these
products are typically required to fit into a tight enclosing
space, which makes disassembly for component recovery a
challenging task.
The low economic value of the material composition, high
rates of material mixing, and low levels of toxic materials
have also discouraged efforts to fully recycle consumer
electronics products. For example, it has been estimated that
3.2 million ton of electronic waste is landfilled each year in
the United States (1). Such high quantities of discarded
electronic products create a risk that hazardous metals such
as lead, mercury, arsenic, and chromium can reach the
environment. The risk is higher for developing countries with
limited environmental controls that may ultimately import
reused or remanufactured electronic products originating
from developed countries (2).
It is well-known that recovery of waste electric and
electronic equipment for reuse or recycling conserves
resources and feedstocks that supply steel, glass, plastic, and
precious metals. Such recycling also avoids air and water
pollution as well as greenhouse gas emissions associated
with materials production and manufacturing. For these
reasons, the number of regulatory and voluntary initiatives
aiming to increase end-of-life (EOL) reuse and recycling is
increasing around the world. Of notable mention is the
impending European Union Directive on Waste Electric and
Electronic Equipment (WEEE), which will require producers
to recycle greater than 50% of consumer products such as
cell phones, coffee makers, and computers by 2006 (3).
With heightened interest in recycling and reuse inevitable
in the coming years, situational factors and uncertainty that
impact the economics of EOL options will require increased
attention and understanding. For example, it is well-known
that product structure, materials, locations of recycling
facilities, applicable regulations, geography, and cultural
context have a major impact on the economic and envi-
ronmental benefits of material recovery. To date, a meth-
odology has yet to be described that can quickly and easily
quantify the trade-offs between reducing environmental
burden and economic costs for optimal EOL strategies,
considering all possible disassembly sequences and com-
ponent fates under multiple situations. Rapid calculation of
this set of solutions is necessary so that the impact of
situational variables such as recycling costs, labor costs, and
transportation distances can be thoroughly understood,
leading to improved EOL decision-making and design for
minimum EOL environmental impact. Moreover, the avail-
ability of a rapid methodology for calculating the optimal
EOL trade-offs would present the opportunity to develop a
concise set of multi-situational EOL strategy graphs for a
given product. Such a concise representation would facilitate
the efforts of consumers and decision-makers to maximize
the environmental benefit of reuse and recycling efforts at
minimum economic cost.
This paper utilizes a multi-objective genetic algorithm
(GA) to establish the optimal set of trade-offs between
environmental impact and cost for EOL strategies among
uncertainty created by situational variables. While the
methodology is presented as an EOL decision-making tool,
it is general in dimensionality such that it could also be used
to simultaneously consider trade-offs related to production
costs, product performance, and life cycle environmental
impact. Therefore the methodology can serve to accelerate
the wider diffusion of green engineering principles such as
those proposed by Anastas and Zimmerman (4) by providing
an approach that engineers can use to quantify and visualize
trade-offs between green engineering outcomes, economic
variables, and design parameters as they arise.
* Corresponding author phone: (734)615-5253; fax: (734)647-3170;
e-mail: skerlos@umich.edu.
Environ. Sci. Technol.
2003,
37,
5303-5313
10.1021/es0345423 CCC: $25.00
�
2003 American Chemical Society
VOL. 37, NO. 23, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5303
Published on Web 10/24/2003 Related Research
Several decision variables must be considered when deter-
mining the maximum environmental benefit that can be
achieved for a given economic cost when a product reaches
EOL. These variables include the extent of disassembly, the
disassembly sequence (if disassembly occurs), and the EOL
fate for removed components as well as the product
remainder not disassembled. A number of previous inves-
tigations have addressed important aspects of this optimiza-
tion problem. For instance, Bras and Emblemsva�g (5)
developed an approach to evaluate the economics of
disassembly under uncertain conditions using activity based
cost modeling. The research investigated the relationship
between specific product designs and the economics of EOL
treatment, while situational variables were evaluated via
Monte Carlo simulation. Rose et al. (6) developed a design-
oriented decision framework applicable to consumer prod-
ucts. In this case, the research focused on technical product
design variables such as expected lifetime and number of
parts, and this information was used to select an appropriate
EOL strategy at the design stage. Similarly, Caudill et al. (7)
investigated the application of life cycle analysis techniques
to facilitate design for multiple product life cycles.
Relative to the research cited above, the field of disas-
sembly sequence planning (DSP) has generally focused less
on product design and EOL management. While material
recovery for environmental benefit has been the main
motivation for DSP, the primary objective of much of the
research has been to maximize the economic returns from
disassembly (8-10) or to maximize the efficiency of disas-
sembly with respect to disassembly time and the number of
components removed (11-14). Also, investigations of sen-
sitivity to situational variables have been rare in the DSP
literature. Recently however, Erdos et al. (16) developed a
DSP sensitivity analysis approach that utilizes a disassembly
AND/OR graph (15) to determine the allowable revenue
reduction for a disassembly action that leaves the optimal
disassembly sequence unaffected. However, the approach
was limited to an evaluation of only the maximum profit
disassembly sequence.
Where environmental objectives have been considered
in DSP, they have typically been treated either implicitly or
as constraints secondary to economic objectives (17-19).
Therefore more relevant to this investigation is the research
performed by Lee et al. (20), which did consider economic
and environmental variables as dual objectives. In this case,
however, the objective function of the optimization problem
was based on a weighted sum of economic and environmental
variables that was used to establish individual component
EOL fates based on only the least cost disassembly sequence.
The methodology was not capable of simultaneous consid-
eration of EOL fate and disassembly sequence and did not
develop Pareto sets of EOL trade-offs. Such Pareto sets were
investigated by Azapagic and Clift (21) in a chemical
engineering application using a linear programming ap-
proach. While effective, the methodology requires multiple
simulations to achieve a single Pareto curve. This could limit
the use of the methodology for exploring a large number of
scenarios as necessary in EOL dec