Page 354 - Materials Chemistry, Second Edition

338 5 Life Cycle Interpretation, Reporting and Critical Review

provided by Huijbregts. 19) For real LCAs, this method, which also neglects figure
margins and only refers to ‘bigger or smaller’, is rarely applied.
The methods described here were supplemented by two others (key issue analysis
and structural analysis) and were tested on the ecoinvent’96 data record. 20) They
have also been implemented in the educational software ‘Chain Management by
Life Cycle Assessment (CMLCA)’. 21)
The issue of data format should rather be part of the inventory (see Section 3.4)
but must be considered in the interpretation. Data formats are often provided
by data bases and software. For newer applications, they include specifications
on statistical distributions of input data. 22) The quality of statistical specifications
must, however, always be critically reflected; standard deviations, for instance, can
be approximated by semi-quantitative procedures.

5.3.3
Non-numerical Methods

Mathematical methods cannot solve problems that result from value choices:
however, increasing the significance of the greenhouse effect or of land use can
have a higher impact on conclusions and recommendations than an increase of
the significance threshold from 0.05 to 0.2. Therefore non-numerical methods are
a fixed part of the interpretation.
Verbal-argumentative interpretation of quantitative and semi-quantitative results
are mainly part of non-numerical methods. Thus, in spite of numerical auxiliary
tools, it is often very difficult to provide numerical limits which, if exceeded, imply
an insure distinction of results for statements like ‘A is better or just as good as
B’. Here it can be verbally referred to the context of earlier experiences or to issues
within the same study. It is the responsibility of the critical review to question
these statements, which are particularly suited for a ‘whitewashing’ of results. It
should, however, not be ignored that seemingly objective figures can be used for
swindling as well; this is just not as evident in times such as ours, bent on figures
and numbers.

5.4
Reporting

The LCA framework ISO 14040 23) states as follows in Chapter 6:

A reporting strategy is an integral part of an LCA. An effective report should
address the different phases of the study. … Results and conclusions of the LCA

19) Huijbregts (1998) and Heijungs and Kleijn (2001).
20) Heijung and Suh (2002) and Heijungs et al. (2005).
21) Chain Management by Life Cycle Assessment (CMLCA); http://www.leidenuniv.nl/cml/ssp/
software/cmlca.
22) Heijungs and Frischknecht (2005).
23) ISO (2006a).