In the heart of Latvia, researchers at Riga Technical University are harnessing the power of artificial intelligence to revolutionize how we assess the environmental impact of our food. Leading this charge is Feofilovs Maksims from the Institute of Energy Systems and Environment, who, along with his team, has developed a novel approach to predict carbon emissions in strawberry production using machine learning. Their work, published in the journal “Vides un Klimata Tehnoloģijas,” which translates to “Environmental and Climate Technologies,” is a significant step forward in making life cycle assessments (LCA) more efficient and data-driven.
Traditionally, LCA methods have relied heavily on human expertise and available data, which can be scarce and inconsistent, especially when comparing different cultivation systems. Maksims and his team aimed to bridge this gap by employing Adaptive Neuro-Fuzzy Inference Systems (ANFIS), a hybrid machine learning technique that combines the strengths of neural networks and fuzzy logic. “We wanted to leverage AI to predict environmental impacts in agriculture, even when data is limited,” Maksims explains. “By using data from greenhouse strawberry production to model open-field systems, we’ve shown that AI can help overcome data scarcity challenges.”
The team trained their model using data generated in MATLAB and validated it against emissions computed using the Ecoinvent 3.10 database and SimaPro software. Among the three fuzzy inference system generation approaches they tested—Fuzzy C-Means (FCM), Subtractive Clustering (SC), and Grid Partitioning (GP)—FCM exhibited the highest accuracy. This methodology not only improves the predictive accuracy of LCAs but also enables more efficient sustainability assessments, which can have significant commercial implications for the energy sector.
As the world grapples with the urgent need to reduce carbon emissions, industries are increasingly seeking ways to quantify and minimize their environmental impact. The energy sector, in particular, is under pressure to adopt sustainable practices and demonstrate their commitment to reducing global warming potential. By providing a more accurate and efficient means of assessing environmental impacts, this AI-driven approach could help energy companies make data-driven decisions that align with their sustainability goals.
Moreover, the ability to use data from one cultivation system to model another could have far-reaching implications for the agricultural sector. As Maksims points out, “This approach could be applied to various crops and cultivation systems, making it a versatile tool for farmers and agribusinesses looking to reduce their carbon footprint.” By extension, this could also benefit the energy sector, which is increasingly investing in bioenergy and other agricultural-based renewable energy sources.
The research conducted by Maksims and his team at Riga Technical University is a testament to the transformative potential of AI in environmental science. As we continue to grapple with the challenges of climate change, such innovative approaches will be crucial in helping industries navigate the complex landscape of sustainability assessments. By embracing AI and machine learning, we can look forward to a future where data-driven decisions pave the way for a more sustainable and environmentally conscious world.
In the words of Maksims, “This is just the beginning. We’re excited to see how this methodology will shape the future of life cycle assessments and contribute to a more sustainable future.” As the energy sector continues to evolve, the insights gained from this research could prove invaluable in driving progress towards a greener, more sustainable future for all.