Repository logo
 
Publication

Life cycle assessment using machine learning

2025-04-14Conference paper Open access
Simple item page
dc.contributor.authorGomes, Sofia
dc.contributor.authorFaria, Brígida Mónica
dc.contributor.authorOliveira, Alexandra Alves
dc.contributor.authorPinto, Edgar
dc.contributor.authorRodrigues, Matilde
dc.contributor.authorVieira, Manuela
dc.contributor.authorFaria, Brigida Monica
dc.contributor.authorOliveira, Alexandra
dc.contributor.authorPinto, Edgar
dc.contributor.authorRodrigues, Matilde
dc.contributor.authorVieira da Silva, Manuela
dc.date.accessioned2025-11-18T12:12:43Z
dc.date.available2025-11-18T12:12:43Z
dc.date.issued2025-04-14
dc.description.abstractLife Cycle Assessment (LCA) is a scientific tool that allows calculating the impact of a product or service on the environment, considering the different phases from planting to transportation, commercialization, consumption, and disposal. (1) LCA requires comprehensive data collection of the inputs and outputs such as raw materials, energy, water, used chemicals and pollutants emissions at each stage of the life cycle. Data is usually obtained from different sources like producers or farmers (primary data), literature reviews, government reports and scientific publications (secondary data) or from associations, non-governmental organizations (NGOs) and international organizations. (2) Data processing and analysis are conducted with the aim of uncovering the resultant environmental impacts. This dissertation, integrated into the project REtail using Technology based on Artificial InteLLigence (RETAILL) (3) aims to apply Machine Learning (ML) techniques to develop surrogate LCA models that can be used to estimate LCA results for new products or services. Both public data and data from the Terras de Felgueiras Cooperative (4) will be used to develop the intended model. By preprocessing and modelling this data, the study aims to provide valuable insights for enhancing sustainability in the production of fresh fruits and vegetables. These insights can guide decision-making and drive continuous improvement in the supply chain. The objective of this study is to develop a ML model that estimates LCA results for new products or services and that translates environmental indicators into measurable impacts on both the environment and human health, specifically focusing on the production process. Another objective is to establish clusters that represent similar environmental performance of producers or products. The first step was to review the existing literature on the subject. To accurately determine emissions from agricultural activities, validated equations from the Agri-footprint 6 methodology (5) were employed. Preliminary analyses and descriptive statistics of variables such as fertilizers, pesticides and fungicides applied on agriculture from public data assessments were conducted using tools like SPSS and RapidMiner. This latter was used to carry out the construction of decision trees and clusters. To evaluate clustering models, certain indices were considered namely the Davies-Bouldin Index and the Calinski-Harabasz Index. Meanwhile, for assessing decision trees, measures such as accuracy rate, F- measure, and confusion matrix are well-known evaluation criteria. Subsequently, the ML model was developed using Python programming language and libraries such as Scikit- learn, Pandas, and SciPy. The analysis of public data reveals results from the cultivation of kiwi, watermelon, citrus, tea, and hazelnut across a total of 865 orchards (6). The results include the development of a tailored ML model for LCA phase that allows. the identification and translation of key environmental indicators into environmental and human health impacts. Furthermore, the clustering results of products and producers enables the observation of patterns in the environmental impact of the production process. Overall, this study contributes to the field of sustainability by providing a framework for integrating ML techniques with life cycle assessment, ultimately leading to more efficient and effective practices in agricultural production. The utilization of validated equations from Agri-footprint 6 enhances the reliability of emissions determination from agriculture, contributing to more accurate assessments of environmental impacts. Ultimately, the goal of an LCA is to support informed decision- making and promote more sustainable practices across industries.por
dc.identifier.citationGomes, S., Faria, B. M., Oliveira, A. A., Pinto, E., Rodrigues, M., & Vieira, M. (2024). Life cycle assessment using machine learning. Proceedings of the 1st Symposium on Biostatistics and Bioinformatics Applied to Health, 5–6. https://recipp.ipp.pt/entities/publication/a634fd4f-6053-47fa-8145-4f876572cba7
dc.identifier.isbn978-989-9045-35-4
dc.identifier.urihttp://hdl.handle.net/10400.22/30956
dc.language.isoeng
dc.peerreviewedn/a
dc.publisherESS | P. PORTO Edições
dc.relation.hasversionhttps://recipp.ipp.pt/entities/publication/a634fd4f-6053-47fa-8145-4f876572cba7
dc.rights.uriN/A
dc.subjectLife cycle assessment
dc.subjectSustainability
dc.subjectMachine learning
dc.subjectData analysis
dc.titleLife cycle assessment using machine learningpor
dc.typeconference paper
dspace.entity.typePublication
oaire.citation.conferenceDate2024-05-03
oaire.citation.conferencePlacePorto
oaire.citation.endPage6
oaire.citation.startPage5
oaire.citation.titleProceedings of the 1st Symposium on Biostatistics and Bioinformatics Applied to Health
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyNameFaria
person.familyNameOliveira
person.familyNamePinto
person.familyNameRodrigues
person.familyNameVieira da Silva
person.givenNameBrigida Monica
person.givenNameAlexandra
person.givenNameEdgar
person.givenNameMatilde
person.givenNameManuela
person.identifierR-000-T1F
person.identifier2545573
person.identifier.ciencia-id0D1F-FB5E-55E4
person.identifier.ciencia-id161A-55D9-C256
person.identifier.ciencia-id271F-B7DF-8FAB
person.identifier.ciencia-id5110-3A70-C3F3
person.identifier.ciencia-id8C1D-7FC0-8460
person.identifier.orcid0000-0003-2102-3407
person.identifier.orcid0000-0001-5872-5504
person.identifier.orcid0000-0002-8021-4783
person.identifier.orcid0000-0001-6175-6934
person.identifier.orcid0000-0003-1256-990X
person.identifier.ridC-6649-2012
person.identifier.ridN-7022-2015
person.identifier.scopus-author-id6506476517
person.identifier.scopus-author-id56340903500
person.identifier.scopus-author-id55485977900
person.identifier.scopus-author-id57449894700
relation.isAuthorOfPublication85832a40-7ef9-431a-be0c-78b45ebbae86
relation.isAuthorOfPublicationd6f940a1-3dba-41d2-9a5e-dc1f313eec07
relation.isAuthorOfPublicationeaf9fc86-1a1c-437f-adee-d28040aa7f2f
relation.isAuthorOfPublication8ddb200d-027c-40ca-8ca4-7e3a5981bcb1
relation.isAuthorOfPublicationfb3cca0a-0dc9-4702-83ae-fbca6b94e1da
relation.isAuthorOfPublication.latestForDiscovery85832a40-7ef9-431a-be0c-78b45ebbae86

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
COM_Sofia Gomes.pdf
Size:
411.47 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
4.03 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

ESS - BBB - Comunicações em eventos científicos