Chemistry Reference
In-Depth Information
encountered here, however, is finding the optimum
improvement strategies and choosing the best alter-
native in a decision environment with a number of
often conflicting objectives. To aid the decision-
making process, an optimisation tool—optimum
LCA performance (OLCAP)—has been developed by
Azapagic and Clift [75]. The OLCAP combines LCA
and process optimisation and defines an optimisation
problem in this context as follows:
other objective function. Therefore, some trade-offs
between objective functions are necessary in order
to reach the preferred optimum solution in a given
situation. Full details of this methodology can be
found in Azapagic and Clift [75] and this is not
elaborated further here. Instead, the value of LCA in
system optimisation is illustrated on a case study of
the Scotch Whisky system.
Minimise
f
(
x
,
y
) = [
f
1
f
2
...
f
p
]
h
(
x
,
y
) = 0
g
(
x
,
y
) < 0
x
Œ
X
Õ
R
n
Case study III: the Scotch whisky system
An LCA study of Scotch grain whisky, used in Scotch
whisky blends with malt whisky, has been carried
out to identify the major environmental impacts
and key stages in the life-cycle with the aim of iden-
tifying options for improving the environmental
performance of the whole system [56]. The system
boundaries include all activities from extraction of
raw materials through crop farming and the manu-
facturing process to the matured product leaving the
factory gate. A flow diagram of the Scotch whisky
life-cycle is shown in Fig. 5.13. The use and product
disposal stages are outside the scope of the study and
therefore are not considered, making this in effect a
'cradle-to-gate' study. The functional unit is defined
as 'operation of the system for one year', which is
equivalent to the total annual output of whisky of
36 million litres of pure alcohol (lpa) and 40 000 t
year
-1
of by-products (animal feed and CO
2
).
As shown in Fig. 5.13, the system is divided into
foreground and background. The foreground com-
prises the manufacturing process itself, i.e. malting
plant, grain distillery and maturation, whereas the
background includes all other activities that supply
material and energy to the foreground, including
the farming subsystem. Cereal grain, either wheat
or maize, is used as the main raw material for the
process. The malting plant provides malted barley,
used in the distillery plant as a source of sacchari-
fication enzymes. The grain distillery subsystem
includes the operations of starch extraction from the
grain by cooking, its conversion to sugars with the
aid of malted barley in the mashing process and
fermentation to metabolise the sugars to produce a
wash of about 7% (v/v) ethanol. Carbon dioxide gas
evolved during fermentation is recovered, purified,
liquefied and sold as a by-product. The spent wash
from distillation contains grain solids that are
processed also as a by-product and sold back to the
agricultural sector as animal feed.
(5.4)
y
Œ
Y
Õ
Z
q
where
f
is a vector of environmental objective func-
tions,
h
(
x
,
y
) = 0 and
g
(
x
,
y
) < 0 are equality and
inequality constraints and
x
and
y
are the vectors
of continuous and integer variables, respectively.
The constraints and objective functions include all
activities from 'cradle to grave' or 'cradle to gate',
depending on the scope of the study. The equality
constraints may be defined by energy and material
balances; the inequality constraints may describe
material availabilities, heat requirements, capacities,
etc. A vector of
n
continuous variables may include
material and energy flows, pressures, compositions,
sizes of unit operations, etc., whereas a vector of
q
integer variables may be represented by alternative
materials or technologies in the system or a number
of trucks for the transport of raw materials. Depend-
ing on the type of constraints and objectives, the
model described by Eqns (5.4) can be linear or
non-linear.
The optimisation problem in the context of LCA is
equivalent to a conventional optimisation model
except that in addition to an economic function it
also involves the environmental objectives, defined
as the burdens or impacts [3,75]. Thus a single objec-
tive optimisation problem usually employed in con-
ventional process optimisation is, in the LCA
context, transformed into a multi-objective one. The
system is optimised simultaneously on a number of
environmental objectives, subject to certain con-
straints encompassing all activities from cradle to
grave. This results in an
n
-dimensional non-inferior
or Pareto surface with a number of optimum solu-
tions for system improvements. By definition, none
of the objective functions on the Pareto surface can
be improved without worsening the value of any
