CyberSEES: Type 1: Preserving the Privacy of Life Cycle Inventory Data in Distributed Provenance Networks

点击翻译按钮获取中文摘要

Life cycle assessment, used to estimate the environmental impacts of products and consumption choices, requires extensive private data about industrial processes. This project will address the question of how claims about the environmental performance of products can be made and evaluated without requiring disclosure of confidential data. First, the data privacy needs associated with life cycle assessment will be articulated in terms of data provenance and scientific workflow publishing. The project will then develop methods for performing life cycle assessment computations that preserve the privacy of input data, while also enabling validation and meaningful interpretation of results. Techniques will be drawn from data anonymization, secure multi-party computation, and homomorphic encryption. The work will consider both scenarios in which private results are shared among collaborators and in which results are made available for public queries. The work will lead to the development of security schemes to address threats to private data used in such computations; algorithms for performing life cycle assessment computations under privacy restrictions; and data anonymization strategies for publication satisfying data providers? privacy requirements.Particularly in the policy arena, but also among the general public, validation of claims about environmental performance in a life cycle perspective is increasingly important. Life cycle assessment enables industrial actors to improve environmental performance and it shows promise for informing sustainability governance. The proposed research will help to normalize the discussion of data privacy, with the intent of fostering a more transparent exchange of information among stakeholders. In turn, this could empower data providers to better maintain their own inventory information, potentially leading to greater availability and quality of life cycle inventory data. The application of privacy preserving computation schemes to environmental informatics has potential to provide businesses with new operational knowledge to guide decision making on sustainability.