Quantifying uncertainty elements in LCI modelling of chemical mixtures used for footwear production

The fashion industry is a fast-growing sector that requires the application of high amounts of chemical substances along the different production processes involved. Consequently, a large degree of pollution is generated, both in terms of ecosystem and human health. Lately, many fashion brands have realized the magnitude of environmental burdens created by their products and have thus committed to transition into more sustainable alternatives. This objective is often supported in the industry by Life Cycle Assessment studies. However, in the long and complex supply chain of fashion products, accurate data is often lacking and difficult to retrieve. For this reason, the chemical substances can be improperly assessed leading to an inaccurate compilation of the Life Cycle Inventory. The purpose of this work is to analyse a specific case study for the production of footwear and quantify quantitative and qualitative sources of uncertainty related to the modelling of chemical substances. These uncertainties are identified in different elements of both foreground and background data systems and are quantified stochastically by randomly sampling a significant number of inventory values within their range of uncertainty. The set of values obtained are used to model the inventory for the production of one functional unit. An impact assessment for global warming potential is performed to evaluate the contribution of the single elements to the total output of the model. Results show that uncertainties in both data systems are substantial, and a wide range of different results can be obtained from the LCA of the same product. The outcome is expected to determine what is the level of confidence in the LCI of industrial products, such as footwear, requiring high use of chemical substances.