Page 114 - Materials Chemistry, Second Edition

98 3 Life Cycle Inventory Analysis

As the allocation requires assumptions, which cannot be strictly and scientifically
deduced, for example, the decision as to whether an allocation per mass, energy
or price is to be made (see Section 3.3.2.3 ‘Diamond paradox’), ISO 14044 82)
recommends the avoidance of allocation by system expansion as the most scientific
solution. Because of the substantially higher effort in practice and depending on
the respective goal definition, multi-output processes are mostly modelled using
defined allocation rules. Most important applications of system expansion can be
found in the assessment of options for the waste industry (see Section 3.3.5) as
well as in the context of open-loop recycling (see Section 3.3.4).

3.3.2.3 Proposed Solutions
For the solution of the problem of allocation different strategies were developed,
of which none, however, completely satisfies. 83) A unique solution for all objectives
and goal definitions is probably impossible.
The rules to be found in ISO 14041 and 14044 and the results of the fundamental
84)
debates at the allocation workshop in Leiden, 1994 are of main concern. Allocation
in LCA has been the topic of several reports that did not always distinguish between
co-products and the related problem of OLR 85) (see Section 3.3.4).
The list below, which can also be used as some sort of a check list, intends to
show the strategies based on ISO 14044, but in a less dogmatical way. The actual
product under investigation by LCA is indicated by A, the co-products by B, C, …
1. Statement whether product B (C, … ) has a performance and is merchantable,
thus represents an economic good. If this is not the case, B is waste (more
precisely: waste for disposal) and no environmental loads are attributed to it.
The underlying logic implies that nobody will conduct a technical process only
in order to produce waste.
2. Examination whether system expansion is possible with justifiable effort. If so,
a scientific solution is possible.
3. Examination whether system reduction is possible: if the unit process is too
large, for example, a whole factory was selected that produces several products
(A, B, C, … ); If so, the large unit process can be divided into smaller ones
considering, for example, a production line, a reactor, a field, and so on, with
only one product under investigation.
This restriction can shift the problem: The separation of a large unit process
into smaller ones implies that the data requirement is much larger and more
differentiated; this means an allocation problem can result in a data problem.
A carefully applied system reduction can also be valued as strictly scientific if
its application does not imply subjective assumptions.

82) ISO, 2006b.
83) Fava et al., 1991; Heintz and Baisn´ ee, 1992; Society of Environmental Toxicology and Chem-
istry – Europe, 1992; Society of Environmental Toxicology and Chemistry (SETAC), 1993;
Boustead, 1994b; Huppes and Schneider, 1994; Ekvall and Tillman, 1997; International Standard
Organization (ISO), 1998a; ISO, 2006b; Curran, 2007, 2008.
84) Huppes and Schneider, 1994; International Standard Organization (ISO), 1998a; ISO, 2006b.
85) Kl¨ opffer, 1996a; Ekvall and Tillman, 1997; Curran, 2007, 2008.