Bogdan Šinik and Aleksandar Tošić

Abstract

Life cycle assessment (LCA) has been established
as the standard method for evaluating environmental impacts of
products, and processes. However, ISO standards depicting the
application of these standards are tailored more towards LCA
practitioners and less on the data acquisition and quality. The
data acquisition process is not very robust, and considerable deci-
sion making and quality control is entrusted to the practitioners.
Taking the lack of incentives from industry participants to submit
quality data, the integrity of LCA databases can be questionable.
Moreover, in some cases, participants may be incentivised to
protect their data from competition. To address these concerns,
data is carefully studied by external experts to verify their
credibility. However, it is not entirely clear how these experts
are chosen, nor how trust is established. In this paper, we apply
a well known method Law of anomalous numbers, commonly
refereed to as Benford’s law in order to test the conformity of
commonly used LCA databases. Our results on testing Ecoinvent,
one of the most widely used LCA databases, show that LCA
data strongly conforms to Benford’s law. Moreover, our analysis
includes 5 additional publicly available LCA databases, which
also conformed with the exception of Bioenergiedat, which is
likely due to the low number of observations. Finally, we tested
individual properties given by Ecoinvent and establish that very
few columns (<5%), which pass the criteria for Benford’s
analysis are non-conforming. Although interesting, these results
call for a more fine-grained analysis as future work.
Index Terms—LCA, Benford’s Law, Anomaly Detection