In recent years, the application of machine learning for virtual sensing has revolutionized the monitoring and management of information. In particular, electrochemical sensors generate large amounts of data, allowing the application of complex machine learning/AI models able to (1) reproduce the measured data and (2) predict and manage faults in the measuring sensor. In this work, data-driven models based on an autoregressive model and an artificial neural network have been identified and used to (i) evaluate sensor redundancy and (ii) predict and manage faults in the context of electrochemical sensors for the measurement of ethanol. The approach shows encouraging results in terms of both performance and sensitivity analyses, allowing for the reconstruction of the values measured by two sensors in a series of six sensors with different dopant levels and to reproduce their values after a fault.

Models are a core element in performing local estimation of the climate change input. In this work, a novel approach to perform a fast downscaling of global temperature anomalies on a regional level is presented. The approach is based on a set of data-driven models linking global temperature anomalies and regional and global emissions to regional temperature anomalies. In particular, due to the limited number of available data, a linear autoregressive structure with exogenous input (ARX) has been considered. To demonstrate their relevance to the existing literature and context, the proposed ARX models have been employed to evaluate the impact of temperature anomalies on rice production in a socially, economically, and climatologically fragile area like Southeast Asia. The results show a significant impact on this region, with estimations strongly in accordance with information presented in the literature from different sources and scientific fields. The work represents a first step towards the development of a fast, data-driven, holistic approach to the climate change impact evaluation problem. The proposed ARX data-driven models reveal a novel and feasible way to downscale global temperature anomalies to regional levels, showing their importance in comprehending global temperature anomalies, emissions, and regional climatic conditions.

This study introduces and evaluates a methodology to define optimal integrated short and long-term air pollution control measures, to support policy formulation by Local Authorities. The approach utilized in this methodology is based on a receding horizon strategy. In this approach, an autoregressive model provides the dynamic characteristics of air quality within a designated time period. The model is established using daily observed data on pollutant concentration, meteorological variables, and estimated emission data in the study area. The model is the core of a model predictive control based on the solution, at each time step, of the resulting optimization problem. The effectiveness of the overall control has been assessed in the context of controlling NO2 concentrations within the city of Milan. The outcomes of the study demonstrate that this control system can serve as a valuable tool to assist Local Authorities in making informed decisions regarding appropriate air quality management strategies.

In this work, a data-driven approach for the identification of a piece-wise linear model for nitrogen oxide daily concentration simulation is presented and applied. The model has been identified by using daily measured concentrations, meteorological variables, and emission levels estimated starting from the results contained in suitable emission databases. We propose an innovative methodology that jointly optimizes clustering and parameter identification. The procedure has been applied considering data from the Milan (Italy) metropolitan area. The methodology has been compared with two state-of-the-art approaches based on a two-step, cluster-based algorithm and on Hammerstein–Wiener models. The results show how, in the presented application, the devised approach ensures better performance with respect to the two literature methods, both in terms of statistical indexes (correlation, normalized mean absolute error) and in terms of problem-specific metrics (hit ratio, false alarm). For this reason, the approach can be considered suitable to be used in the definition of optimal emission control strategies.

Climate change mitigation is crucial for the health and sustainability of our planet. This paper presents a novel system utilizing the Model Predictive Control technique (MPC) to optimize greenhouse gas emissions reductions. Through a receding horizon approach, the system dynamically adjusts carbon dioxide, methane, and other greenhouse gases emissions to maintain temperature anomalies within safe limits. The system operates in two phases: first, applying MPC to minimize temperature anomalies by optimizing emissions from 2025 to 2100; second, embedding an artificial neural network trained to capture the relationship between policy decisions and greenhouse gas emissions into an optimization process that identifies the policy set required to achieve a desired emissions pathway. The proposed two-phase approach successfully constrains temperature anomalies near the 1.5°C threshold by 2100, limits mid-century overshoot, and computes optimal emission pathways that can be translated into actionable policy measures. Consequently, the presented tool supports policymakers by providing data-driven, scientifically validated decisions, integrating advanced control strategies and neural network modeling to offer a comprehensive solution for sustainable climate change mitigation.

The relationship between climate change and migration has gained significant attention in recent years, particularly regarding its impact on vulnerable regions. This study proposes a novel modeling approach to climate-induced migration by systematically comparing traditional statistical models and machine learning techniques. While most existing research in this field relies on linear models or case-specific analyses, our work explicitly models complex non-linear and multidimensional relationships among environmental and socio-economic drivers of migration. By comparing autoregressive, polynomial, and logarithmic models with a Random Forest model, we demonstrate the added value of machine learning in capturing complex patterns that conventional models often fail to detect. Furthermore, we analyze how variables such as temperature anomalies, the Human Development Index (HDI), water stress, and the agricultural sector’s contribution to the Gross Domestic Product influence migration. This framework is applied to data-scarce and climate-sensitive regions, namely North Africa, Sub-Saharan Africa, and Southeast Asia, where robust modeling remains both challenging and urgently needed. Model performance was systematically evaluated through multiple metrics, including correlation, Mean Error (ME), Mean Absolute Error (MAE), and their normalized forms (NME and NMAE). Results show that while simpler models exhibit limited explanatory power, the Random Forest model substantially improves predictive accuracy, achieving correlation = 0.81, MAE = 0.95, NME =–0.009, and NMAE = 0.25. These findings demonstrate the model’s effectiveness in capturing migration dynamics, and also provides a replicable methodology that can support targeted adaptation strategies, effective migration management, and climate policies, particularly relevant for data-scarce regions of the Global South.