New approach for data quality management in carbon footprint assessment of supply chain processes: the case of OVS Spa

Climate change is one of most pressing issue of our global society. In 2015 the Paris Agreement set the goal to keep the rise in mean global temperature to well below 2 °C above pre-industrial levels, and preferably limit the increase to 1.5 °C, because this would substantially reduce the effects of climate change. To do so, greenhouse gases (GHG) emissions must be reduced drastically across all sectors of industry and human activities in general. In 2018 the fashion industry contributed to at least 4% of the global GHG emissions and so has a relevant role in the present and future emission reductions that are required to face the climate crisis. To guide GHGs emissions reduction it is essential to have a proper estimation of such emissions in terms of representativeness of the actual process, namely an estimation based on high quality data. This work is meant to improve the modelling of the garment manufacturing stages in the carbon footprint model of clothing items, by using data quality assessment as an a priori selection method for emission factors from databases, resulting in a new reversed approach to data quality management for a carbon footprint assessment in the context of a complex supply chain. This approach allows DQA to be coupled since the beginning with the modelling phase, with the main advantage of guiding it from the start on a path towards appropriateness and optimal process representativeness. Given its significant advantages, this new approach to data quality management is the main result of this study, even beyond the calculated emission factors and the actual improvement of the model. This approach is applied to OVS, a large Italian fashion retailer, which uses an indicator of the GHG emissions generated by the production of each one of its items that is based on a carbon footprint model. The method applied to the OVS case returns a remarkable improvement of the manufacturing stages modelling in terms of process representativeness and overall data quality.