How to address data quality in the solid waste domain

In life cycle assessment (LCA) data quality assessment is done to evaluate the reliability and properly interpret the results of the study. For many years there has been awareness about the importance of data quality in LCA,... [ view full abstract ]

In life cycle assessment (LCA) data quality assessment is done to evaluate the reliability and properly interpret the results of the study. For many years there has been awareness about the importance of data quality in LCA, e.g. Kennedy et al. (1996a) who emphasized the importance of data quality due to the large quantity and number of different sources of input data in LCA. Still today, the evaluation of data quality is a relevant issue. In waste-LCA, a review found that less than half of 222 published studies included a critical discussion on used data.

The pedigree matrix approach (Weidema and Wesnæs, 1996) is accepted as the main tool for systematic data quality assessment in LCA. The approach which was adopted from the Numeral, Unit, Spread, Assessment and Pedigree (NUSAP) system, developed by Funtowicz and Ravetz (1990), and is used to give a structured scoring of the different data used to build a life cycle inventory for a LCA. Common data quality indicators are representativeness, completeness, precision, consistency and reliability of the data; whereas less common data quality indicators are repeatability/reproducibility and process review. Indicators can be assessed at flow level, process level and/or at system level (entire LCI).  

The pedigree matrix have also been criticized for the degree of subjectivity when assigning scores to the quality indicators. Weidema (1998) stated that use of the pedigree approach must involve subjective judgments of the scoring criteria in order to obtain correct assessment of the data at hand. It is though found that this rarely is done studies applying the pedigree matrix, and the drawback from modifications is that comparisons across studies become impossible if there is not a degree of consistency in the approach.

In order  to find a way to improve this for the solid waste management domain the following actions were taken 1) described the process of adapting generic data quality indicators to a specific domain in an operational, transparent and systematic manner, 2) developed data quality indicators, which are adapted to the domain of solid waste management, and 3) applied this to a case example.