Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation
Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation
Life Cycle Assessment and Its Application To Process Selection, Design and Optimisation
4.67H
2
O), boric acid (H
3
BO
3
), anhydrous borax (Na
2
B
4
O
7
),
and anhydrous boric acid (B
2
O
3
). A simplied LCA ow
diagram of the system, showing the distinction between the
foreground and background subsystems, is shown in
Fig. 15. After extraction and crushing, the boron minerals,
borax and kernite, are transported to the plant and dissolved
in water to produce 5 and 10 mol borates. Boric acid (BA) is
produced in a separate plant, by reacting kernite ore with
sulphuric acid. Anhydrous borax (AB) and anhydrous boric
acid (ABA) are made in high-temperature furnaces from
5 mol borate and BA, respectively. All products are then
either packed or shipped in bulk. Electric energy and steam
for the system are provided by the on-site natural gas
cogeneration facility with additional steam generated in
the steam plant. All activities from the extraction of raw
materials to the production of the boron products and
materials used are included in the system; however, the
use and disposal phases of the products are not considered in
this study (`cradle-to-gate' approach). The functional unit
was dened as total yearly production of the boron products
of 1,062,000 t/year.
The objective of this work was to optimise the system on
environmental and economic performance to identify a
range of possibilities for minimising total environmental
impacts from the system, while maximising production
subject to total product demand and keeping the production
costs to their minimum. The optimisation model also
included several process alternatives to identify the BPEO.
These included different product dryers (uid, tray and
rotary), conveyors instead of trucks in the mine, and the
generation of additional steam in the Cogeneration instead
of in the Steam plant. To demonstrate the approach, the
system model dened by Eqs. (1)(4), was optimised on
three objective functions: global warming potential (GWP),
total production (P) and production costs (C). The three-
dimensional non-inferior surface ABCD, generated in a
series of multi-objective optimisations, is shown in
Fig. 16. Depending on the position on the non-inferior
surface, the optimum solutions offer different options for
improvements and BPEOs. For instance, Point A represents
the minimum of the cost-objective function; however, the
Fig. 15. LCA flow diagram of the boron system (from Azapagic and Clift
[110]).
Fig. 16. Non-inferior surface for the boron system (from Azapagic [20]) (C
*
, GWP
*
and P
*
optimum values obtained in single-objective optimisation; C,
GWP, P optimum values obtained in multi-objective optimisation).
A. Azapagic / Chemical Engineering Journal 73 (1999) 121 15
production is at the minimum and GWP is 31% above its
optimumvalue. The BPEOat this point includes transport in
the mine by trucks and steam production in the Steam plant.
The KuhnTucker multipliers
3
at this solution suggest that if
a further decrease in the GWP objective by one tonne is
required, a cost increase of 95 would be incurred for
constant product output; similarly, if the production were
to increase by one tonne, the resulting increase in the costs
would be equal to 34 for constant GWP. The KuhnTucker
multipliers therefore determine the marginal costs of redu-
cing GWP and increasing production, respectively.
By moving from Point A along the non-inferior curve for
constant GWP, both costs and production increase, to reach
their maximum feasible values at point B. The Cost function
here is 4% above its optimum value. If Production is
increased by one tonne, 300 of the Costs objective have
to be given up. Similarly, one tonne change in GWP is
associated with a cost change of 100. At this solution, the
BPEO is dened by steam generation in the Steam plant and
the preferred transportation means in the mine is by con-
veyors.
If, however, the system were to be operated at Point C,
GWP would be 3.3% above its optimum value obtained in
single objective optimisation. Production would be at the
minimum, and the costs would increase by 14%. An
improvement in GWP of one tonne would worsen the values
of the Costs objective by 36, while a tonne increase in P
would result in 100 increase in the costs. At this point, 93%
of the steam is generated by the Cogeneration plant and the
rest is produced in the Steam plant. The conveyors still
remain the best transport option in the mine.
Furthermore, if, for example, Point D were to be chosen
as the best compromise solution, then for the same value of
GWP as at Point C, production would reach the maximum;
however, costs would have to increase by 17%. It can be
noted here that both GWP and Production exhibit similar
effect on Costs: a decrease in GWP by one tonne increases
Costs by 3,600, while increasing production by one tonne,
increases the costs by 3,500. At this solution, the best
practicable environmental option is dened by truck trans-
port in the mine and steam production in the Cogeneration
plant.
These results demonstrate how optimum solutions and
therefore BPEO change with the operating state of the
system. The same analysis can be done for other points
on the non-inferior surface which are all optimal in the
Pareto sense. By trading-off the values of different objec-
tives at these points, decision makers can select any solution
on the surface, depending on how much of one objective
they are prepared to give up in order to gain in another. The
value of multi-objective optimisation in the context of LCA,
therefore, lies in offering a range of choices for environ-
mental and economic improvements of the system and so
enabling preferences to be identied after analysing all the
trade-offs among objectives.
Although the discussion in this section has mainly
focused on optimisation of the existing processes, a similar
approach can be applied for the design of new processes.
The use of LCA and multi-objective optimisation in the
process design is reviewed in the next section.
4.3. LCA for process design
One of the newly emerging applications of LCA is in
product and process design. This has resulted in the devel-
opment of a new LCA-related tool Life Cycle Product/
Process Design (LCPD). The LCPD methodology is still in
its infancy and published literature is scant (e.g.
[92,94,138]).
Although the methodologies for life cycle product and
process design are similar, the following discussion will
focus on Life Cycle Process Design. As outlined in Fig. 17,
environmental considerations are incorporated at an early
stage of the design, alongside the more traditional technical
and economic criteria [139]. LCA is used throughout the
design process, initially on a reference process. The con-
ventional system boundary is extended to include the life
cycles of different technologies and raw materials, all the
way from extraction of primary resources through to pro-
duction. This enables a quantitative comparison of different
technological routes for production of the same set of raw
materials as well as an assessment of different raw materi-
als. It may be noted that the same rigorous owsheeting
procedures used in conventional design can directly be
linked with the environmental analysis described here.
Furthermore, as Product Stewardship initiatives are seeking
to build alliances between manufacturers and suppliers
3
Kuhn-Tucker multipliers indicate how much of one objective function
has to be given up in order to improve the value of the other; for more
detail see e.g. [128].
Fig. 17. General methodological framework for Life Cycle Process
Design (from Azapagic [139]).
16 A. Azapagic / Chemical Engineering Journal 73 (1999) 121
[140], it is also desirable to include the supply chain within
the design framework. This enables identication of best
suppliers, in terms of their product quality, environmental,
health, safety and other performance criteria. For instance,
one of the guiding principles in choosing the supplier could
be their compliance with ISO 9000 or 14000 or perhaps an
international environmental label of their product. More-
over, the design process must include compliance with
relevant legislation, such as health and safety regulations
and environmental emissions limits. Finally, customer and
consumer requirements in terms of the specications and
performance of the product can also be addressed during
process design. Once all of the criteria of interest have been
identied, a multi-objective optimisation model of the
general form Eqs. (1)(4) can be dened. The system is
then optimised on a number of objectives, dened by
environmental burdens or impacts and socio-economic
functions, subject to the constraints on material and energy
balances, productive capacities, technical, legislative and
other requirements. As a result, a plethora of non-inferior
solutions is obtained, enabling a quantitative evaluation of
options for environmental, technical, economic and other
improvements of the process system. This whole procedure
is dynamic with a continuous exchange of information
within and outside the design team to explore systematically
the possibilities for improvements.
Therefore, LCPD offers a potential for technological
innovation in the process concept and structure through
the selection of the best technologies and raw materials over
the whole cycle. As already discussed, this can be of
particular importance if placed within the context of EMAS
and ISO 14000 EMS, as well as the IPPC Directive, which
require companies to have a full knowledge of the environ-
mental consequences of their actions, both on- and off-site.
Furthermore, as the `polluter pays' and `producer respon-
sibility' initiatives are starting to force manufacturers to
reduce waste at source and manage the waste associated
with their processes, LCPD can provide a powerful frame-
work for the design of cleaner processes which are envir-
onmentally benign and economically protable.
Similar approaches have been proposed by Kniel et al.
[100], Pistikopoulos et al. [102], Stewart and Petrie [104],
and others [103,105]. Pistikopoulos et al. [102,103,105], for
instance, use a so-called Minimum Environmental Impact
(MEI) methodology which embeds LCA principles within a
formal process optimisation framework. This approach
extends the existing waste minimisation design techniques
by providing a more complete description of the environ-
mental impacts of the process. The authors [102,103,105]
use an example of the design of a vinyl chloride monomer
(VCM) plant to show that the `zero emission' target may not
always be the best environmental policy and that an optimal
degree of abatement may be preferable instead. This is
illustrated in Fig. 18 which reveals that there is a minimum
mass of dichloroethane (DCE) discharged beyond which the
global environmental impact increases due to the trade-offs
in waste generation over the whole life cycle between
system inputs and outputs [102]. This example indicates
that from an environmental point of view, minimisation of
the output emissions only, as normally carried out in con-
ventional system optimisation, can in fact lead to sub-
optimal solutions. Furthermore, it also emphasizes the
importance of multi-objective optimisation in which all
functions have to be considered simultaneously if global
environmental improvements are to be achieved. Targeting
for minimum `global' impacts may also result in less
expensive plant operation, as was the case in this study.
5. Conclusions
There is a growing need to move away from narrow
denitions and concepts in environmental system manage-
ment. Life Cycle Assessment offers a systematic way to
incorporate the entire material and energy supply chain into
strategic planning and policy development. This is demon-
strated by an increasing number of applications of LCA,
both by industry and governments. Some of the newly
emerging applications of LCA reviewed in this paper are
in process selection, design and optimisation.
The application of LCA to process selection shows that
the Best Practicable Option (BPEO) that benets the envir-
onment as a whole must be identied and chosen in the LCA
context. The case studies conrm that the choice of BPEO is
dependent on the boundaries, operating state of the system
and on the background economic system in which it oper-
ates. Some of the examples indicate that recycling and
recovery of pollutants may not always be the best option
environmentally and that the BPEO has to be determined on
a case-by-case basis.
LCA can also be coupled with multi-objective optimisa-
tion techniques to provide a powerful tool for process design
and optimisation. A newly emerging Life Cycle Process
Design (LCPD) tool offers a potential for technological
Fig. 18. Effect of the degree of abatement on water discharges from the
VCM plant (from Pistikopoulos et al. [102]).
A. Azapagic / Chemical Engineering Journal 73 (1999) 121 17
innovation in process concept and structure through selec-
tion of best material and process alternatives over the
whole life cycle. This approach provides a robust frame-
work for process design by simultaneously optimising on
environmental, technical, economic and other criteria.
Multi-objective optimisation in this context serves to
identify a number of Pareto-optimum options for improved
design and operation throughout the whole life cycle. This
approach provides a potentially powerful decision making
tool which may help process industries identify sustainable
options for the future.
Acknowledgements
I wish to express my thanks to Prof. Roland Clift for
introducing me to LCA some years ago and for his invalu-
able contributions to some of the works quoted here. I am
also grateful for his and Prof. John Smith's suggestions for
improvements to this paper.
References
[1] ISO 14001, Environmental Management Systems Specification
with Guidance for Use, HMSO, London, 1996.
[2] Council Regulation (EEC) No. 1836/93 of 29 June 1993 Allowing
Voluntary Participation by companies in the Industrial Sector in a
Community Eco-Management and Audit Scheme 1993. Official
Journal of the European Communities, No. L 168, 10 July, HMSO,
1993, London, pp. 118.
[3] Council Directive 91/61/EC 1996. Concerning Integrated Pollution
Prevention and Control, Official Journal of the European Commu-
nities, No. L 257, 10 October, HMSO, London, 1996.
[4] Department of the Environment, Transport and the Regions, U.K.
Implementation of EC Directive 96/61 on Integrated Pollution
Prevention and Control, Second Consultation Paper, DETR,
London, 1998.
[5] R. Clift, A.J. Longley, Introduction to clean technology, in: R.C.
Kirkwood, A.J. Longley (Eds.), Clean Technology and the
Environment, Chap. 6, Blackie Academic and Professional,
Glasgow, 1994.
[6] R. Clift, Clean technology The idea and practice, J. Chem. Tech.
Biotech. 68 (1997) 347350.
[7] G. Allen, R. Clift, J.F. Davidson (Eds.), Clean Technology:
The Idea and the Practice. Philosophical Transactions: Mathema-
tical, Physical and Engineering, 355, Royal Society, 1997,
p. 1728.
[8] R. Clift, Engineering for the environment: The new model engineer
and her role, Process Safety and Environ. Protection 76(B2) (1998)
151160.
[9] I. Boustead, The Milk Bottle, Open University Press, Milton
Keynes, 1972.
[10] B. Hannon, System Energy and Recycling: A Study of the
Beverage Industry, Center for Advanced Computation, University
of Illinois, Urbana, IL, 1972.
[11] G. Sundstrom, Investigation of the energy requirements from raw
materials to garbage treatment for 4 Swedish Beer Packaging
Alternatives, Report for Rigello Park AB, Sweden, 1973.
[12] R.G. Hunt, W.E. Franklin, Resources and Environmental Profile
Analysis of 9 Beverage container Alternatives. Contract (68-01-
1848), U.S. EPA, Washington, DC, 1974.
[13] E. Barber, K. Fanganas, F. Iannazzi, R. Shamel, K. Halloack, H.
Frost, Selected Characteristics of Disposable and Reusable
Napkins, (PB-275 222/8), Arthur D. Little, Cambridge MA, 1977.
[14] R.U. Ayres, Process classification for the industrial material sector,
Technical Report, United Nations Statistical Office, New York,
1978.
[15] M.P. Lundolm, G. Sundstrom, Resource and Environmental Impact
of Tetra Brik Carton and Refillable and Non-Refillabel Glass
Bottles, AB Tetra Pak, Malmo, 1985.
[16] I. Boustead, Environmental Impact of the Major Beverage
Packaging Systems U.K. Data 1986 in Response to the EEC
Directive 85/339, INCPEN, London, 1989, pp. 14.
[17] ISO/DIS 14040, Environmental Management Life Cycle Assess-
ment Part 1: Principles and Framework, 1997.
[18] J. Fava, R. Dennison, B. Jones, M.A. Curran, B. Vigon, S. Selke, J.
Barnum (Eds.), A Technical Framework for Life-Cycle Assess-
ment, SETAC and SETAC Foundation for Environmental Educa-
tion, Washington, DC, 1991.
[19] F. Consoli, D. Allen, I. Boustead, J. Fava, W. Franklin, A.A. Jensen,
N. de Oude, R. Parrish, R. Perriman, D. Postlethwaite, B. Quay, J.
Seguin, B. Vigon (eds.), Guidelines for Life-Cycle Assessment: A
`Code of Practice', SETAC, Brussels, 1993.
[20] A. Azapagic, Environmental system analysis: The application of
linear programming to Life Cycle Assessment, Ph.D dissertation,
University of Surrey, 1996.
[21] ISO/TC 207/SC 5/WG 2 N 112, ISO/DIS 14041: Environmental
Management Life Cycle Assessment Part 2: Goal and Scope
Definition and Life Cycle Inventory Analysis, Voting draft, 1997.
[22] ISO/TC 207/SC 5/ N 112, ISO/DIS 14042.3: Environmental
Management Life Cycle Assessment Part 3: Life Cycle Impact
Assessment, Committee draft, 1998.
[23] ISO/TC 207/SC 5/ N 112, ISO/DIS 14043: Environmental
Management Life Cycle Assessment Part 4: Life Cycle
Interpretation, Voting draft, 1998.
[24] A. Azapagic, R. Clift, Allocation of environmental burdens by
whole-system modelling The use of linear programming. In: G.
Huppes, F. Schneider, (Eds.), Allocation in LCA, SETAC, Brussels,
1994, pp. 5460.
[25] R. Clift, R. Frischknecht, G. Huppes, A.-M. Tillman, B. Weidema,
(Eds.), Toward a coherent approach to Life Cycle Inventory
Analysis, Report of the Working Group on Inventory Enhancement,
SETAC-Europe, Brussels, 1998.
[26] H.A. Udo de Haes, J.-F. Bensahel, R. Clift, C.R. Fussler, R.
Griesshammer, A.A. Jensen, Guidelines for the application of Life
Cycle Assessment in the EU ecolabelling programme, European
Commission, DG XI-A-2, Brussels, 1994.
[27] H. Hallay, R. Pfriem, O