Estimating Missing Unit Process Data in Life Cycle Assessment Using a Similarity-Based Approach

Abstract

In life cycle assessment (LCA), collecting unit process data from the empirical sources (i.e., meter readings, operation logs/journals) is often costly and time-consuming. We propose a new computational approach to estimate missing unit process data solely relying on limited known data based on a similarity-based link prediction method. The intuition is that similar processes in a unit process network tend to have similar material/energy inputs and waste/emission outputs. We use the ecoinvent 3.1 unit process data sets to test our method in four steps: (1) dividing the data sets into a training set and a test set; (2) randomly removing certain numbers of data in the test set indicated as missing; (3) using similarity-weighted means of various numbers of most similar processes in the training set to estimate the missing data in the test set; and (4) comparing estimated data with the original values to determine the performance of the estimation. The results show that missing data can be accurately estimated when less than 5% data are missing in one process. The estimation performance decreases as the percentage of missing data increases. This study provides a new approach to compile unit process data and demonstrates a promising potential of using computational approaches for LCA data compilation.

Methodological review and detailed guidance for the life cycle interpretation phase

“…The use of input–output modeling for building the LCI can also help identify potentially important elements of the system that have been left out (Mattila, 2018). Alternative techniques have also been proposed to estimate influence of missing information such as the use of algorithms, like the similarity‐based approach by Hou, Cai, and Xu (2018) or the FineChem approach for chemical production proposed by Wernet, Hellweg, Fischer, Papadokonstantakis, and Hungerbühler (2008) and Wernet, Papadokonstantakis, Hellweg, and Hungerbühler (2009).…”

Exploring machine learning techniques to predict deforestation to enhance the decision‐making of road construction projects

“…Concerning the LCA methodology, the use of ML is yet to become popular, although there is an increasing number of studies that use this technique to build life‐cycle inventories (Liao et al., 2019), predict environmental impacts (Nabavi‐Pelesaraei et al., 2018), estimate characterization factors (Marvuglia et al., 2015; Hou et al., 2020), or to perform data imputation in environmental databases (Hou et al., 2018). For instance, ML has been used to predict crop production based on production variables to spatialize life‐cycle impacts (Lee et al., 2020).…”

Application of Life Cycle Assessment and Machine Learning for High-Throughput Screening of Green Chemical Substitutes

“…and outputs such as wastes. 19 Nevertheless, the framework was not suitable for the complex pharmaceuticals' manufacturing process because of numerous unknown LCIA data.…”

“…Moreover, LCIA data for pharmaceutical processes were always rarely available because of the use of fine chemicals with complex molecular structures, which increased the challenges to carry out the LCA calculation of pharmaceuticals production and optimize the chemicals’ production process. , Recently a similarity-based link prediction approach has been developed to predict LCIA data in a given chemical process according to the similarity theory, that is, similar processes tend to share similar inputs (e.g., materials, energy, etc.) and outputs such as wastes . Nevertheless, the framework was not suitable for the complex pharmaceuticals’ manufacturing process because of numerous unknown LCIA data.…”