Improving the integrated hybrid LCA in the upstream scope 3 emissions inventory analysis

Purpose

The protocols of carbon footprints generally define three scopes for different greenhouse gas (GHG) emissions levels. The most important carbon footprint emissions source comes from upstream indirect emissions of scope 3 for products that do not consume energy during their use phase. Upstream scope 3 GHG inventory can usually be analyzed through input–output or hybrid LCA analysis. The economic input–output life cycle analysis (EIO-LCA) and the hybrid LCA model have been widely used for this purpose. However, a cutoff error exists in the hybrid model, and the lack of a truncation criterion between process and IO inventory may lead to a high level of uncertainty in the hybrid model. This study attempts to improve the problem of cutoff uncertainty in hybrid LCA and proposes a method to minimize the cutoff uncertainty.

Methods

The way to improve the cutoff uncertainty could follow two steps. First, through the IO inventory analysis of EIO-LCA, we can define the emissions by various tiers of product components. The IO inventory indicator can provide a definitive criterion for the process inventory of the hybrid model. Second, we connect the process- and IO-LCI according to the IO inventory result. The advantage of the process inventory is that it provides detailed manufacturing information on the target while the IO encompasses a complete system boundary. For improvements, the process inventory can catch the most important process of the GHG emissions, and the IO inventory could compensate for the remainder of the incomplete system inventory.

Results and discussion

In this case study, the printed circuit board production process is used to evaluate the efficiency of the improved method. The threshold M was set to 70 in this case study, and the IO inventory provides the remaining 30 %. For the integrated hybrid model, the tier 3 process inventory takes only 64 % while the incorporation of the proposed method can include 92 % of the total emissions, which shows the cutoff uncertainty can be reduced through the improvement.

Conclusions

This study provides a clear guideline for process and IO cutoff criteria, which can help the truncation uncertainty. When higher precision is required, process LCI will need to play an important role, and thus, a higher M value should be set. In this situation, the emissions from IO-LCI would be smaller than the emissions from the process LCI. The appropriate solution would attain a comfortable balance between data accuracy and time and labor consumption.