Scholarlink

<p>Conducting Life Cycle Assessments (LCAs) for machines and tools in manufacturing is often time-intensive and hampered by difficulties in accessing suitable datasets. It is, nevertheless, important to consider the life cycle impacts of the machines and tools in a manufacturing process to identify hot spots, evaluate the relative importance of their impacts and develop strategies for further environmental improvements. This core novelty of this study is in the investigation of whether heterogeneous life cycle assessment (LCA) inventory and impact data for manufacturing machines can be meaningfully clustered, and whether the resulting groupings can be interpreted and operationalised as generalised proxy life cycle inventory datasets. It further evaluates the practical usefulness of these cluster-derived proxy datasets for supporting screening-level LCA and early-stage sustainability decision-making in manufacturing contexts. Three proxy LCA categories were developed, capturing broad similarities between machines while retaining sufficient detail for high-level sustainability assessments. The resulting machine proxy data provides a practical tool for streamlining LCA decision-making, allowing practitioners to estimate life cycle impacts from production to end-of-life even in the absence of detailed datasets. Validation against individual ecoinvent datasets showed that these generalised categories produce reasonably accurate approximations within typical uncertainty ranges, supporting exploratory analyses and screening applications. However, they are not intended to replace full LCAs for specific machines where precise assessment is required. Future work could enhance the proxy datasets by incorporating real-time operational data and regional variations, potentially using machine learning to refine impact estimates dynamically. Industrial integration of these datasets, such as in digital twin models or automated LCA platforms, would enable rapid, scalable sustainability assessments, supporting more informed decision-making in machine selection, procurement, and operational planning.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deriving proxy life cycle assessment datasets for manufacturing machines through data clustering

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月6日

查看全文
  • Nathan Dodd,
  • Lenny Koh,
  • Rachael Rothman

摘要

Conducting Life Cycle Assessments (LCAs) for machines and tools in manufacturing is often time-intensive and hampered by difficulties in accessing suitable datasets. It is, nevertheless, important to consider the life cycle impacts of the machines and tools in a manufacturing process to identify hot spots, evaluate the relative importance of their impacts and develop strategies for further environmental improvements. This core novelty of this study is in the investigation of whether heterogeneous life cycle assessment (LCA) inventory and impact data for manufacturing machines can be meaningfully clustered, and whether the resulting groupings can be interpreted and operationalised as generalised proxy life cycle inventory datasets. It further evaluates the practical usefulness of these cluster-derived proxy datasets for supporting screening-level LCA and early-stage sustainability decision-making in manufacturing contexts. Three proxy LCA categories were developed, capturing broad similarities between machines while retaining sufficient detail for high-level sustainability assessments. The resulting machine proxy data provides a practical tool for streamlining LCA decision-making, allowing practitioners to estimate life cycle impacts from production to end-of-life even in the absence of detailed datasets. Validation against individual ecoinvent datasets showed that these generalised categories produce reasonably accurate approximations within typical uncertainty ranges, supporting exploratory analyses and screening applications. However, they are not intended to replace full LCAs for specific machines where precise assessment is required. Future work could enhance the proxy datasets by incorporating real-time operational data and regional variations, potentially using machine learning to refine impact estimates dynamically. Industrial integration of these datasets, such as in digital twin models or automated LCA platforms, would enable rapid, scalable sustainability assessments, supporting more informed decision-making in machine selection, procurement, and operational planning.