Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production

Artificial Intelligence (AI) is transforming traditional methods reliant on human knowledge by introducing machine learning techniques, which offer effective solutions to complex challenges. An example of such a case is the evaluation of the environmental impacts of products throughout their life cy...

Full description

Saved in:
Bibliographic Details
Main Authors: Feofilovs Maksims, Zaeemi Majid, Cappelli Andrea, Romagnoli Francesco
Format: Article
Language:English
Published: Sciendo 2025-01-01
Series:Environmental and Climate Technologies
Subjects:
Online Access:https://doi.org/10.2478/rtuect-2025-0017
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849225016820891648
author Feofilovs Maksims
Zaeemi Majid
Cappelli Andrea
Romagnoli Francesco
author_facet Feofilovs Maksims
Zaeemi Majid
Cappelli Andrea
Romagnoli Francesco
author_sort Feofilovs Maksims
collection DOAJ
description Artificial Intelligence (AI) is transforming traditional methods reliant on human knowledge by introducing machine learning techniques, which offer effective solutions to complex challenges. An example of such a case is the evaluation of the environmental impacts of products throughout their life cycle. This study bridges the gap in life cycle assessment (LCA) by leveraging AI to predict environmental impacts in agriculture, specifically by using LCA data from one cultivation system to model another. We employed Adaptive Neuro-Fuzzy Inference Systems (ANFIS) to predict CO2 equivalent emissions for open-field strawberry production, utilizing greenhouse strawberry data. The novelty lies in combining machine learning with LCA to address data scarcity and improve predictive accuracy in agricultural impact assessments. The model was trained with data generated in MATLAB and validated against emissions computed using the Ecoinvent 3.10 database and SimaPro software. Among the three fuzzy inference system (FIS) generation approaches - Fuzzy C-Means (FCM), Subtractive Clustering (SC), and Grid Partitioning (GP) FCM exhibited the highest the accuracy. This methodology showcases AI’s potential to transform LCA, enabling more efficient, data-driven sustainability assessments.
format Article
id doaj-art-40d6fec57a12486f951490fb0a177602
institution Kabale University
issn 2255-8837
language English
publishDate 2025-01-01
publisher Sciendo
record_format Article
series Environmental and Climate Technologies
spelling doaj-art-40d6fec57a12486f951490fb0a1776022025-08-25T06:12:10ZengSciendoEnvironmental and Climate Technologies2255-88372025-01-0129124325810.2478/rtuect-2025-0017Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry ProductionFeofilovs Maksims0Zaeemi Majid1Cappelli Andrea2Romagnoli Francesco3Institute of Energy Systems and Environment, Riga Technical University, Azenes iela 12/1, Riga, LV-1048, Latvia2Bioeconomy in Transition Research Group, IDEA, Unitelma Sapienza - University of Rome, Viale Regina Elena, 295, 00161Rome, Italy3Department of Chemical Engineering Materials Environment (DICMA), Faculty of Civil and Industrial Engineering, Sapienza University of Rome, Via Eudossiana 18, Rome, 00184, ItalyInstitute of Energy Systems and Environment, Riga Technical University, Azenes iela 12/1, Riga, LV-1048, LatviaArtificial Intelligence (AI) is transforming traditional methods reliant on human knowledge by introducing machine learning techniques, which offer effective solutions to complex challenges. An example of such a case is the evaluation of the environmental impacts of products throughout their life cycle. This study bridges the gap in life cycle assessment (LCA) by leveraging AI to predict environmental impacts in agriculture, specifically by using LCA data from one cultivation system to model another. We employed Adaptive Neuro-Fuzzy Inference Systems (ANFIS) to predict CO2 equivalent emissions for open-field strawberry production, utilizing greenhouse strawberry data. The novelty lies in combining machine learning with LCA to address data scarcity and improve predictive accuracy in agricultural impact assessments. The model was trained with data generated in MATLAB and validated against emissions computed using the Ecoinvent 3.10 database and SimaPro software. Among the three fuzzy inference system (FIS) generation approaches - Fuzzy C-Means (FCM), Subtractive Clustering (SC), and Grid Partitioning (GP) FCM exhibited the highest the accuracy. This methodology showcases AI’s potential to transform LCA, enabling more efficient, data-driven sustainability assessments.https://doi.org/10.2478/rtuect-2025-0017anfisartificial intelligencecarbon footprintglobal warming potentialsustainability
spellingShingle Feofilovs Maksims
Zaeemi Majid
Cappelli Andrea
Romagnoli Francesco
Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production
Environmental and Climate Technologies
anfis
artificial intelligence
carbon footprint
global warming potential
sustainability
title Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production
title_full Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production
title_fullStr Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production
title_full_unstemmed Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production
title_short Machine Learning Approach for Predicting Environmental Impact: A Neuro-Fuzzy Model for Life Cycle Impact Assessment of Strawberry Production
title_sort machine learning approach for predicting environmental impact a neuro fuzzy model for life cycle impact assessment of strawberry production
topic anfis
artificial intelligence
carbon footprint
global warming potential
sustainability
url https://doi.org/10.2478/rtuect-2025-0017
work_keys_str_mv AT feofilovsmaksims machinelearningapproachforpredictingenvironmentalimpactaneurofuzzymodelforlifecycleimpactassessmentofstrawberryproduction
AT zaeemimajid machinelearningapproachforpredictingenvironmentalimpactaneurofuzzymodelforlifecycleimpactassessmentofstrawberryproduction
AT cappelliandrea machinelearningapproachforpredictingenvironmentalimpactaneurofuzzymodelforlifecycleimpactassessmentofstrawberryproduction
AT romagnolifrancesco machinelearningapproachforpredictingenvironmentalimpactaneurofuzzymodelforlifecycleimpactassessmentofstrawberryproduction