WE.1.E || Green-Lean-Digital
Cornago, Simone; Tan, Yee Shee; Ramakrishna, Seeram; Low, Jonathan Sze Choong
The Life Cycle Inventory (LCI) is the most time and resource consuming phase within Life Cycle Assessment (LCA). Data collection often relies on aggregated data, such as monthly averages, or on primary data from short periods of time that might not be fully representative of the underlying process. However, with the growing availability of interconnected sensors through the Industrial Internet of Things (IIoT), it has become more feasible to continuously and automatically gather information on the variability of the primary data through a dynamic LCI. Since the IIoT allows a cheap collection of a continuous stream of data, reducing a set of values for every time-step into an LCI average, albeit more precise, might be a waste of knowledge. Hotspot analysis is commonly used to identify the sources of the main drivers of potential environmental impacts. In this study, we tackle a problem of hotspot analysis in the traditional LCA, which is that it considers a static assessment and cannot account for a dynamic LCI. Here we show a new methodology to implement a dynamic hotspot analysis to fully harness the information of a dynamic LCI, based on the matrix approach of the Brightway2 framework. Although there are several works in the literature to implement dynamic LCAs, they do not include a dynamic hotspot analysis. This would allow a more detailed understanding for which material flows, energy flows or processes might be responsible for the highest shares of the LCA impacts. Such a feature would be useful to inform Life Cycle Management decision making. Indeed, a dynamic LCI enables the recognition of the potential temporal variability of the sensorized processes and that of the related LCA impacts. The dynamic hotspot analysis could then help to highlight the LCA impacts drivers that, if successfully managed, could bear the most effective LCA impacts reductions.