Forest science and artificial intelligence applied to more resilient forests and more sustainable landscapes.
We are an interdisciplinary research group bringing together specialists in forest sciences and artificial intelligence, with the aim of developing innovative solutions for adaptive forest management.
We recognize that today's challenges — climate change, social transformation, and technological change — require new approaches to forest management, capable of integrating ecological knowledge, digital technologies, and advanced data analysis.
Interdisciplinary
Forest sciences
& Artificial Intelligence
LiDAR
Cutting-edge technology
ML
Predictive models
Research centre
Work developed within the Research Centre in Asset Management.
Turn complex environmental data into useful information for decision support, promoting adaptive management approaches that strengthen the response capacity of the forestry sector.
Decisions based on data and quantitative science
Capacity to respond to climate uncertainty
Contribution to better prepared rural communities
Three-dimensional data for detailed large-scale forest structure.
Low-cost aerial monitoring, adaptable to different scales and terrains.
Accessible tools for field data collection by any user.
These technologies make it possible to collect detailed information on forest structure and condition, enabling applications that assess forest stands, monitor their development, and identify damage caused by disturbances or extreme weather events.
Models that understand and anticipate the evolution of forest ecosystems over time.
Decision-support tools for multifunctional forest management across multiple spatial scales.
Small owners who manage their forests and need accessible, intuitive tools.
Professionals working in planning and managing forest resources in the field.
Local actors with a key role in implementing sustainable practices in the territory.
The challenge: These actors face limits in access to technical information, analytical tools, and decision-support systems. We aim to develop accessible, intuitive, user-oriented solutions that facilitate monitoring, planning, and decision-making.
We support a future in which forests are more resilient, ecologically balanced, and socially valued, playing a central role in territorial sustainability and community well-being.
We promote adaptive forest management, guided by the principles of observe, anticipate, and adapt, applied across multiple spatial and temporal scales and supported by scientific evidence, data, and accessible digital innovation.
We believe in collaborative artificial intelligence, where technology works together with people, expanding their ability to understand complex systems and support more informed decisions about land management.
We support democratizing access to new digital tools by making them available to technicians, managers, and territorial agents, strengthening analysis, planning, and decision-making capacities in forest management.
We embrace open science principles, promoting open-source code, methodological transparency, and public data availability whenever possible, as a way to strengthen trust in the knowledge produced and encourage collaboration.
Inspired by the new concept of silviculture proposed by Achim and collaborators — increasingly guided by data, science, and systematic knowledge.
Adaptive Cycle
Continuous data collection
Predictive models
Adjustment of interventions
This permanent adaptive cycle guides decision-making and allows management practices to be adjusted in response to climate uncertainty and society's new demands.
Structure integrated data collection and analysis with prospective modeling and AI as decision support.
→ Reconfiguration of technical rationale
Implement, monitor, and adjust solutions in real-world contexts, integrating ecological and territorial variability.
→ Operational adaptive management models
Democratize access to advanced analytical tools and strengthen territorial technical skills.
→ Evidence-based professional culture
Demonstrate feasibility and replicate across multiple contexts and territories.
→ New forest management paradigm
| Dimension | Purpose | Intended Transformation |
|---|---|---|
| Scientific and Digital Infrastructure | Structure integrated data collection and analysis with prospective modeling and AI as decision support | Reconfiguration of the technical rationale of forest management, replacing empirical decisions with decisions informed by quantitative science |
| Demonstration and Adaptive Management | Implement, monitor, and adjust solutions in real-world contexts, integrating ecological and territorial variability | Consolidation of operational adaptive management models capable of responding to climate uncertainty and disturbances |
| Training and Transformation of Decision Culture | Democratize access to advanced analytical tools and strengthen territorial technical skills | Transition to a professional culture based on evidence, technical autonomy, and informed accountability |
| Structural Impact and Scalability | Demonstrate feasibility and replicate across multiple contexts | Scientific and social legitimization of a new forest management paradigm |
Some of the scientific work supporting the project's approach and demonstrating the application of computer vision, artificial intelligence, spatial analysis, and predictive models in forest and industrial contexts.
14 publications gathered
More information about the projectCastro, I.; Salas-González, R.; Fidalgo, B.; Farinha, J.T.; Mendes, M. Optimising Forest Management Using Multi-Objective Genetic Algorithms. Sustainability 2024, 16, 10655. https://doi.org/10.3390/su162310655
Henriques, L., Farinha, T. & Mendes, M. Fault Detection and Prediction for a Wood Chip Screw Conveyor. Eksploatacja i Niezawodność - Maintenance and Reliability, 26(3), 2024.
Falcão, D., Reis, F., Farinha, J., Lavado, N., & Mendes, M. Fault Detection in Industrial Equipment through Analysis of Time Series Stationarity. Algorithms, 17(10), 455, 2024.
Mateus, B. C., Mendes, M., Torres Farinha, J., Marques Cardoso, A., Assis, R., & Soltanali, H. Improved GRU prediction of paper pulp press variables using different pre-processing methods. Production & Manufacturing Research, 11(1), 2023.
Martins, A., Mateus, B., Fonseca, I., Farinha, J. T., Rodrigues, J., Mendes, M., & Cardoso, A. M. Predicting the Health Status of a Pulp Press Based on Deep Neural Networks and Hidden Markov Models. Energies, 16(6), 2651, 2023.
Rodrigues, J.A., Farinha, J.T., Cardoso, A.M., Mendes, M., Mateus, R. Prediction of Sensor Values in Paper Pulp Industry Using Neural Networks. In Proceedings of IncoME-VI and TEPEN 2021. Mechanisms and Machine Science, vol 117. Springer, Cham, 2023.
Mateus, B., Mendes, M., Farinha, J.T., Martins, A.B., Cardoso, A.M. Data Analysis for Predictive Maintenance Using Time Series and Deep Learning Models - A Case Study in a Pulp Paper Industry. In Proceedings of IncoME-VI and TEPEN 2021. Mechanisms and Machine Science, vol 117. Springer, Cham, 2023.
Guimarães, A.; Valério, M.; Fidalgo, B.; Salas-Gonzalez, R.; Pereira, C.; Mendes, M. Cork Oak Production Estimation Using a Mask R-CNN. Energies 2022, 15, 9593.
André Guimarães, Beatriz Fidalgo, Carlos Pereira, Raúl Salas-Gonzalez, and Mateus Mendes. Estimativa da produção de cortiça através de um método não destrutivo da árvore usando Mask R-CNN. In 1º Congresso Nacional em Engenharia e Gestão de Ativos, Instituto Superior de Engenharia de Coimbra, Portugal, 26/27 May 2022.
Rodrigues, J. A., Farinha, J. T., Mendes, M., Mateus, R. J. G., & Cardoso, A. J. M. Comparison of Different Features and Neural Networks for Predicting Industrial Paper Press Condition. Energies, 15(17), 6308, 2022.
Salas GR e B Fidalgo. Impacto de agentes de distúrbio nos serviços dos ecossistemas em povoamentos de pinheiro bravo na Serra da Lousã. Geografia, Riscos e Proteção Civil. Livro Homenagem ao Professor Doutor Luciano Lourenço. Volume 2; 213-224, 2021.
Coelho, J.; Fidalgo, B.; Crisóstomo, M.M.; Salas-González, R.; Coimbra, A.P.; Mendes, M. Non-Destructive Fast Estimation of Tree Stem Height and Volume Using Image Processing. Symmetry 2021, 13, 374.
Mateus, B. C., Mendes, M., Farinha, J. T., & Cardoso, A. M. Anticipating Future Behavior of an Industrial Press Using LSTM Networks. Applied Sciences, 11(13), 6101, 2021.
Mateus, B. C., Mendes, M., Farinha, J. T., Assis, R., & Cardoso, A. M. Comparing LSTM and GRU Models to Predict the Condition of a Pulp Paper Press. Energies, 14(21), 6958, 2021.
