Interpretation of drivers associated with the occurrence of floods through explainable artificial intelligence

doi:10.22167/2675-6528-202602619

Article

September 29, 2026

Interpretation of drivers associated with the occurrence of floods through explainable artificial intelligence

Interpretation of Drivers Associated with the Occurrence of Floods through Explainable Artificial Intelligence

Abderraman Róger de Amorim Brandão; Wagner Wilson Bortoletto

DOI: 10.22167/2675-6528-202602619

Article derived from a Course Conclusion Work (TCC), with content based on the student’s original work and adapted to the editorial format of the E&S Magazine with the support of the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege for textual synthesis and organization.

Summary

Flood events in Brazil have intensified, but their factors are still difficult to generalize due to regional contrasts and nonlinear interactions. The study characterized the spatio-temporal distribution of flood events in Brazilian river basins and evaluated the ability of a Random Forest (RF) model to detect days associated with these events based on antecedent hydrometeorological conditions. The research covered 735 basins in the period from 1980 to 2010. Events were identified in daily flow series, using a basin-specific threshold (50th percentile of annual maximums) and a minimum interval of 10 days between peaks. The RF model was trained with dynamic predictors and interpreted with TreeSHAP. The results indicated seasonal differences between biomes, with a higher concentration of events in certain months in the Amazon, Cerrado, Caatinga, and Atlantic Forest, and a more uniform distribution in the Pampa. In the test set, the model achieved a PR-AUC of 0.41, ROC-AUC of 0.96, F1 of 0.35, and balanced accuracy of 0.86. SHAP analyses highlighted the accumulated climatic water balance (P – ETP) and accumulated precipitation in windows of 14 to 60 days as the main predictors. These patterns were compatible with the influence of antecedent conditions but did not establish causal relationships.

Keywords: Antecedent conditions; Large sample hydrology; Random Forest; Hydrological simulation; XAI.

1. Introduction

In Brazil, extreme hydrological events, such as floods, have become more frequent and intense in recent decades, affecting different regions of the country. In the Amazon, for example, an expansion of floodable areas has been observed (Fleischmann et al., 2023; Silva et al., 2023), while the South recorded extreme floods in 2023 and 2024 (Alvalá et al., 2024; Collischonn et al., 2025), with 340 municipalities in a state of emergency (ANA, 2024). Similar occurrences have been documented in the Northeast (Seigerman et al., 2024; Santos et al., 2024), the Center-West (Marengo et al., 2022), and the Southeast. In the state of São Paulo, an average of 73 events per year were recorded in the 2010s (Brandão et al., 2026).

The intensification of these extreme hydrological events poses significant challenges, especially in a country with great dependence on water resources, where 49% of electricity is of hydroelectric origin, irrigation accounts for 50% of water withdrawals, and urban supply for 23.34% (ANA, 2024). Globally, disasters associated with weather, climate, and water caused more than two million deaths and losses of US$ 3.64 trillion between 1970 and 2019, with floods accounting for 44% of these events (WMO, 2021). In Brazil, reported losses in this period totaled US$ 41.7 billion (WMO, 2021). More recently, between 2020 and 2024, 771.75 thousand people were reported affected by floods, with material damages of R$ 3.68 billion and public and private losses exceeding R$ 9 billion (Brasil, 2026). These data reinforce the urgency in understanding the factors associated with floods.

The occurrence of floods results from the complex interaction between atmospheric processes and the physical characteristics of river basins (Sharma et al., 2018; Livneh et al., 2024). The isolated analysis of precipitation, by itself, is often insufficient to represent the contribution of antecedent conditions and nonlinear interactions that modulate the hydrological response of basins. The difficulty in generalizing flood drivers on a national scale, due to strong regional contrasts and nonlinear interactions between climate and hydrological response, represents a significant gap in knowledge and predictive capability.

In this context, machine learning (ML) emerges as a promising tool to represent complex relationships between multiple predictors and support the investigation of patterns associated with extremes (Reichstein et al., 2019; Nearing et al., 2024). However, ML models, such as tree ensembles, can present decisions that are difficult to inspect (Baehrens et al., 2010; Lundberg & Lee, 2017). Explainable artificial intelligence (XAI), through techniques like SHapley Additive exPlanations (SHAP), allows for local and global interpretations of predictions (Lundberg & Lee, 2017), helping to relate input features to model outputs more transparently and to engage with physical knowledge. This approach is fundamental to overcoming the limitations of traditional physics-based models, which, despite being well-founded, do not always produce accurate predictions and are computationally intensive and difficult to generalize (Nearing et al., 2021; Zounemat-Kermani et al., 2021).

Given the increasing frequency and impact of floods in Brazil and the complexity of their drivers, this study is justified by the need to deepen the understanding of the spatio-temporal patterns of these events and by the application of advanced machine learning and XAI methods to identify their predictors. The objective of this study was to characterize the spatio-temporal distribution of fluvial flood events, defined by a flow exceedance criterion, in Brazilian river basins. The ability of a Random Forest (RF) model to detect days associated with these events from hydrometeorological variables was also evaluated. Finally, we sought to identify, through SHAP, the predictors that contributed most to the classifications, with an emphasis on the hydrological interpretation of the learned patterns.

2. Material and Methods

The methodology of this study was structured in three main stages: data organization and identification of flood events, development of the machine learning model, and finally, evaluation and interpretation of the results. This flow guided the research from the preparation of the database to the final analysis of the model.

The study covered 735 Brazilian hydrographic basins, distributed across different biomes and hydroclimatic conditions, with a predominance of tropical climate A (Beck et al., 2018). The CABra dataset — Catchment Attributes for Brazil (Almagro et al., 2021) provided daily flow series from the National Water and Basic Sanitation Agency (ANA) and hydroclimatic information from various sources, including physical and hydrological attributes, for the period from 1980 to 2010.

River flood events were identified from daily discharge series. An event was defined when the daily discharge exceeded a specific threshold for each basin, established as the 50th percentile of annual maxima (Brunner et al., 2020; Götte & Brunner, 2024). A minimum interval of ten days between successive peaks was adopted for event separation (Diederen et al., 2019). For the daily classifier, event days were labeled as 1 and others as 0.

The development of the machine learning model used a Random Forest (RF) classifier (Breiman, 2001; James et al., 2023). The model was built with 39 dynamic hydroclimatic predictors, calculated to represent the antecedent conditions to floods. These predictors included accumulated precipitation over different time windows, sums of actual and potential evapotranspiration, daily climatic water balance and its accumulations, as well as moving averages of temperature, humidity, wind speed, and solar radiation. The month and day of the year were included for seasonality.

From the initial 39 predictors, 15 with the lowest global contributions, measured by the average of absolute SHAP values in an exploratory model, were excluded, resulting in 24 variables for the final model. The RF hyperparameters were evaluated by RandomizedSearchCV (Bergstra & Bengio, 2012), testing 50 random configurations with cross-validation in three partitions and grouping by basin (GroupKFold). The average precision (AP) was the selection criterion.

The final configuration of the RF model consisted of 603 trees, entropy criterion, maximum depth of 40, square root of the number of attributes per split, minimum of 11 samples per leaf and 20 samples for node splitting, with `balanced_subsample` weighting and `bootstrap` sampling. To handle class imbalance, the training set included all positive examples and a random sampling of negatives per basin, limited to up to 20 negatives per positive, an `undersample` strategy.

The separation between the training and testing sets was performed by watershed, using the GroupShuffleSplit method. 588 watersheds (80%) were allocated for model development and 147 watersheds (20%) for evaluation on watersheds not used in training, preventing days from the same watershed from appearing simultaneously in both sets.

The contribution of the predictors was interpreted with SHapley Additive exPlanations (SHAP), an explainable artificial intelligence (XAI) approach (Lundberg & Lee, 2017; Lundberg et al., 2020). TreeSHAP was used to calculate Shapley values, employing the `tree-path-dependent` approach. SHAP values were calculated from a stratified sample of the 147 test basins, with up to 200 event days and 400 non-event days per basin, and a global limit of 5,000 observations. The global importance of each predictor was calculated by the mean of its absolute SHAP values.

The classifier’s performance was evaluated on the test set using metrics appropriate for binary classification in an imbalanced context: ROC-AUC, PR-AUC, F1-score, and balanced accuracy (BA). PR-AUC was the primary metric due to the low prevalence of positive days (Saito & Rehmsmeier, 2015). Threshold-dependent metrics were calculated with a value of 0.5 for converting probabilities into binary classes. Precision, sensitivity, F1-score, and balanced accuracy were defined according to their standard formulations in classification problems.

3. Results and Discussion

The investigation of river flood patterns in Brazilian watersheds revealed a complex dynamic, influenced by spatio-temporal factors and antecedent hydrometeorological conditions. The initial analysis of flood events, identified by a threshold of excess daily flow, demonstrated an episodic nature. It was observed that most years in the period from 1980 to 2010 presented low or moderate event counts, around 5,000 occurrences, punctuated by peak years with approximately 20,000 records in some biomes. This temporal variability underscores the non-stationarity of extreme events, a crucial aspect for water resource planning and management in Brazil.

The spatial distribution of flood events highlighted the heterogeneity of the monitoring network and the intrinsic characteristics of each biome. The Atlantic Forest and the Cerrado concentrated the largest number of basins in the sample, with 371 and 219 basins, respectively, followed by the Amazon (78 basins) and Caatinga (50 basins). The Pampa, with 16 basins, and the Pantanal, with only one, showed less representation. This disparity in data coverage is an important factor to consider in the interpretation of the results, as it may limit the generalization of patterns to less represented biomes, such as the Pantanal.

The seasonality of flood events showed distinct patterns among biomes. In the Amazon, Cerrado, Caatinga, and Atlantic Forest, events were concentrated in the months of December to April or May, with a notable reduction during the winter period. This behavior is consistent with regional differences in flood seasonality, as pointed out by Bartiko et al. (2019). In contrast, the Pampa exhibited a more uniform distribution of events throughout the year, which aligns with the absence of a well-defined dry season in this region. In the Cerrado, the frequency of events decreased more rapidly throughout the year, which can be attributed to both the greater depth of the rocky basement and the strong seasonal concentration of rainfall.

The analysis of peak flows revealed that the Amazon presented a greater dispersion of values, while the Pantanal, with its single basin in the sample, registered high peaks. It is fundamental to consider that these differences in peak flows are influenced by the drainage area and the size of the rivers, and do not necessarily indicate variations in the relative intensity of floods between the biomes. The interpretation of this data must, therefore, take into account the geomorphological and hydrological characteristics of each basin, which modulate the hydrological response to precipitation events.

Regarding the predictive capability of the Random Forest (RF) model to detect days associated with flood events, the overall performance on the external test set was evaluated by specific metrics for imbalanced binary classification problems. The prevalence of positive days in the test set was 1.49%, indicating the rarity of the events. In this scenario, the area under the Precision-Recall curve (PR-AUC) reached 0.41, a value approximately 27.4 times higher than the random baseline, demonstrating the model’s relevance for rare classes (Saito & Rehmsmeier, 2015).

Complementarily, the area under the Receiver Operating Characteristic curve (ROC-AUC) was 0.96, suggesting a good relative class ordering capability by the model. However, the F1 score of 0.35 and the balanced accuracy of 0.86, calculated with a threshold of 0.5, indicate that although the model is effective in discrimination, the balance between precision and sensitivity still presents limitations at this threshold. This performance difference is expected, given the highly unbalanced and temporally explicit nature of the flood classification problem, compared to spatial susceptibility studies that often use balanced samples.

In comparative terms, the PR-AUC of 0.41 obtained in this study is at the lower limit of values reported in other applications of RF models for classifying floodable areas or flood susceptibility. For example, studies in Kunming (Ren et al., 2024), Salzburg (Nachappa et al., 2020), the Peneus basin (Tepetidis et al., 2025), and the eastern Brazilian Amazon (Amaral et al., 2024) reported AUC values between 0.87 and 0.99. However, this comparison should be made with caution, as most of these works focused on the spatial classification of flooded points or pixels, often with balanced samples, unlike the present study which addresses the detection of rare events in a temporal context.

The spatial variability of the model’s performance, measured by PR-AUC per basin, was predominantly moderate, with most values concentrated between 0.3 and 0.6. It was observed that some of the highest PR-AUC values occurred in the Northeast, while intermediate and low performances were distributed across other regions. The relationship between the event rate per basin and PR-AUC revealed a positive trend: basins with a higher frequency of events tended to show higher PR-AUC values. In contrast, basins with very rare events exhibited greater dispersion and lower metric values, reflecting the intrinsic difficulty of learning in minority classes and the greater instability of the metric in this context.

The interpretation of the contribution of predictors through SHapley Additive exPlanations (SHAP) was crucial for understanding the model’s behavior. Global importance analyses highlighted that the greatest contributions were associated with predictors accumulated over windows of 14 to 60 days. In particular, the accumulated climatic water balance (P – ETP) and accumulated precipitation were the most influential predictors. The climatic water balance, which represents the difference between precipitation and potential evapotranspirative demand, acts as an indirect indicator of antecedent conditions, although it does not directly measure soil moisture or water storage in the basin.

This pattern of predictor importance is consistent with the hydrological literature, which emphasizes the joint contribution of precipitation and antecedent conditions to flood generation (Sharma et al., 2018; Livneh et al., 2024). Basin aridity has also been recognized as a central parameter in modulating hydrological processes (Almagro et al., 2024; Brandão et al., 2025; Correa et al., 2025). The results obtained refine this interpretation by highlighting that the “basin memory”, represented by accumulated hydrometeorological variables in different time windows, is a significant component in flood prediction.

Thermoclimatic variables, such as the 30-day average minimum temperature, showed a less monotonic relationship with SHAP values, suggesting their influence is more indirect, reflecting effects of seasonality, precipitation, and evapotranspiration, as well as regional differences. This indicates that, while precipitation and water balance exert a more direct influence on flood occurrence, thermoclimatic variables act mainly as seasonal and regional conditioners, modulating the hydrological response of the basins.

The spatial distribution of the local importance of predictors also revealed regional contrasts. The 30- and 60-day accumulated precipitation and P – ETp were more important in the Northeast, Center-East, and Southeast. In contrast, the 7- and 14-day accumulated precipitation and P – ETp had greater relevance in parts of the South and Southeast. The 30-day average minimum temperature, in turn, showed a greater contribution in the South region. These regional patterns highlight the complexity of flood drivers and the need for approaches adapted to the specificities of each Brazilian region.

SHAP dependency plots for the most important variables revealed non-linear relationships between predictor values and their contributions to the model output. Low values of P – PET and accumulated precipitation in 7, 14, 30, and 60-day windows were generally associated with negative SHAP values, meaning they reduced the predicted probability of event occurrence. As these accumulations increased, contributions tended to become positive, raising the probability of flooding. This transition in contribution sign, where SHAP equals zero, represents a shift in the predictor’s influence relative to the model’s baseline, rather than a physical flood threshold.

This behavior is hydrologically coherent, as floods are the result of both recent rainfall and antecedent soil moisture conditions in the basin. The visual separation between points classified as events (concentrated in regions with positive SHAP) and non-events (more frequent in ranges with negative SHAP) reinforces the high discriminative power of these variables, especially for the 30- and 60-day accumulated P – ETP and precipitation. This suggests that the RF model effectively uses hydrological memory on weekly to monthly scales to identify patterns associated with flood occurrence.

Despite the relevance of the identified patterns, some limitations must be considered. The definition of flood events, based on the 50th percentile of annual maxima, influences the results, although previous studies indicate that the choice of thresholds affects absolute values more than dominant relative patterns (Brunner and Chartier-Rescan, 2024; Götte and Brunner, 2024). The absence of static attributes in the final model limits the direct representation of basin structural controls, which could enrich the understanding of hydrological processes.

The balancing strategy, which used all positive examples and a random sampling of negative examples per basin, was adequate to reduce information leakage between basins, but did not completely eliminate the challenges associated with the strong spatial heterogeneity of performance and the low prevalence of events in some test basins. Furthermore, the SHAP-based interpretation, while useful for describing model behavior, should not be interpreted as direct causal evidence, as correlations between predictors can influence the distribution of contributions.

No formal tests for temporal or spatial autocorrelation were performed, meaning that the possible dependence between consecutive days and between neighboring or hydrologically connected basins remains a limitation. The heterogeneity of the sample, especially the presence of only one basin in the Pantanal, limits the extrapolation of observed patterns to the entire territory of each biome. These aspects highlight the need for future research that addresses these dependencies, incorporates static attributes, and evaluates explicit forecasting horizons.

In summary, this study characterized the spatio-temporal distribution of riverine flood events in Brazil and demonstrated the capability of the Random Forest model in detecting them based on antecedent hydrometeorological conditions. The results indicated seasonal differences between biomes and highlighted the accumulated climatic water balance and accumulated precipitation in 14- to 60-day windows as the most important predictors. These findings are consistent with the influence of basin hydrological memory on the classifications, providing valuable insights into the patterns associated with floods, despite the inherent limitations of the hydrological system’s complexity and the nature of the data.

4. Conclusion

This study aimed to characterize the spatio-temporal distribution of riverine flood events in Brazilian river basins and to evaluate the capacity of a Random Forest (RF) model to detect them from hydrometeorological variables, in addition to identifying the most relevant predictors through SHAP. It was found that flood events exhibited an episodic nature and distinct seasonal patterns among biomes, with a higher concentration in certain months in the Amazon, Cerrado, Caatinga, and Atlantic Forest, and a more uniform distribution in the Pampa. The RF model demonstrated good discriminative capacity on the test set, with PR-AUC of 0.41 and ROC-AUC of 0.96, indicating its relevance for detecting rare classes. SHAP analyses highlighted the accumulated climatic water balance (P – ETP) and accumulated precipitation in 14- to 60-day windows as the most influential predictors, revealing non-linear and hydrologically coherent relationships that emphasize basin memory. The main contribution lies in the in-depth understanding of patterns associated with floods on a national scale and in demonstrating the applicability of explainable artificial intelligence methods to interpret the influence of antecedent conditions.

Despite the advances, the study presented limitations, such as the dependence of the results on the definition of the flood threshold and the absence of static attributes in the final model, which could represent structural controls of the basins. The strong spatial heterogeneity of the model’s performance and the low prevalence of events in some test basins also posed challenges, and the interpretation based on SHAP should not be considered direct causal evidence. For future studies, it is suggested to evaluate temporal and spatial autocorrelation, incorporate static attributes, and explore different definitions of events and prediction horizons. It is also recommended to compare the RF model with Long Short-Term Memory (LSTM) networks to assess the gain of sequential representation, maintaining data consistency and evaluation metrics.

Bibliographic References

Agência Nacional De Águas E Saneamento Básico – ANA (Brasil). Conjuntura dos recursos hídricos no Brasil 2024

Alvalá, S. R. C., Ribeiro, D. F., Marengo, J. A., Seluchi, M. E., Gonçalves, D. A., da Silva, L. A., … & Saito, S. M. (2024). Analysis of the hydrological disaster occurred in the state of Rio Grande do Sul, Brazil in September 2023: vulnerabilities and risk management capabilities. International Journal of Disaster Risk Reduction, 110, 104645.

Collischonn, W., Fan, F. M., Possantti, I., Dornelles, F., Paiva, R., Medeiros, M. S., … & Piccilli, D. G. A. (2025). The exceptional hydrological disaster of April-May 2024 in southern Brazil. RBRH, 30, e1.

Fleischmann, A. S., Papa, F., Hamilton, S. K., Fassoni-Andrade, A., Wongchuig, S., Espinoza, J. C., … & Collischonn, W. (2023). Increased floodplain inundation in the Amazon since 1980. Environmental Research Letters, 18(3), 034024.

Silva et al., 2023 [Full reference not found in the original document]

Article originating from the Final Course Project of the Specialization in Data Science and Analytics of the MBA USP/Esalq

To learn more about the course, click here and access the MBX Academy platform

Most recent

Economy, Investments And Banking

October 23, 2026

Financial Education, Behaviors and Biases in Financial Decision Making

October 22, 2026

Challenges and Possibilities of the First School Year of Children with ASD for Teaching Work

October 22, 2026

Consumer Journey in the Decision to Purchase Infoproducts in the Digital Environment

October 22, 2026

Psychosocial Risks and Mental Health at Work in a Banking Institution

October 22, 2026

Digital Observatory of the Amazon: Data Science Applied to the Analysis of Social, Economic, Environmental, and Health Indicators

October 22, 2026

Continuous Learning as a Factor of Organizational Growth in the Perception of Collaborators

October 22, 2026

Statistical Regression in Microgrids for Operational Performance Estimation: Campusgrid Case

October 22, 2026

Forecast of Aggregate Default in the Brazilian Financial System by Sarimax Model with Macroeconomic Variables

October 22, 2026

Stakeholder Engagement Factors in International Technology Transfer to a Public Hospital

Personal financial management represents a growing challenge, influenced by cognitive biases that affect decision-making, even in individuals with high knowledge. The present study analyzed the relationship between the level of financial education and the occurrence of behavioral biases in the financial decision-making of adults residing in Brazil. An applied, quantitative, and descriptive-correlational field survey was conducted by administering a structured questionnaire to 163 respondents. The instrument included sociodemographic variables, financial knowledge and habits, and hypothetical situations to identify cognitive biases. The results indicated that higher levels of financial education were associated with better financial control, saving, and investment practices, with significant correlations with income and schooling. However, the presence of behavioral biases (anchoring, loss aversion, scarcity, optimism/negativity, present bias, status quo, overconfidence, and mental accounting) persisted even among individuals with higher schooling and income, with no statistically significant association with the analyzed sociodemographic variables. It was concluded that financial education, although essential for financial organization and control, was not sufficient to neutralize the influence of cognitive biases. The incorporation of behavioral approaches into financial education programs is suggested.

Keywords: Financial behavior; Behavioral economics; Personal finance; Cognitive heuristics; Cognitive biases.

October 22, 2026

Challenges and Possibilities of the First School Year of Children with ASD for Teaching Work

The inclusion of children with Autism Spectrum Disorder (ASD) in Early Childhood Education presents significant challenges, given the difficulty of adapting to new environments and the intensity of stimuli. The objective was to analyze the teaching performance in Early Childhood Education in the initial schooling process of children with ASD. A case study was conducted, using narrative interview as a data production technique. The results revealed that the teaching performance was characterized by mediation and affection practices, with the flexibility of routine and the use of playful resources as essential strategies for the child’s development and sensory organization. The importance of the school support professional for the success of the pedagogical work was also highlighted. The child’s integral development was promoted through intentional teaching actions that prioritized environmental adaptation, strengthening of bonds, and attention to the specific needs of ASD. As a contribution, a guide of pedagogical strategies aimed at welcoming children with ASD in public daycare centers was developed.

Keywords: Early Childhood Education; School Inclusion; Teacher Narratives; Inclusive Pedagogical Practices; Autism Spectrum Disorder.

October 22, 2026

Consumer Journey in the Decision to Purchase Infoproducts in the Digital Environment

The expansion of e-commerce and technological transformation have driven the need to understand consumer behavior regarding infoproducts, whose perceived experience has become crucial for professional development. The study aimed to analyze the factors that influenced the purchase decision of infoproducts in the digital environment, considering the stages of the consumer journey. An applied, quantitative, and descriptive research was conducted through a field survey with 100 consumers from Serra Gaúcha. Data collection occurred via a structured questionnaire, using non-probabilistic sampling. The results revealed a qualified audience with strong “mobile-first” behavior. It was found that the availability of detailed information and social approval were the main attraction triggers in the pre-purchase phase. In the transactional and subsequent stages, the minimization of operational friction, the agility of immediate access, and system security proved crucial for conversion and satisfaction, although gaps in human support efficiency were identified. Retention was based on the trust generated by positive past experiences, overcoming the price factor. It was concluded that the infoproduct consumer journey operates as an integrated system, requiring alignment between technological infrastructure, content curation, and support humanization to ensure engagement.

Keywords: E-commerce; Consumer behavior; Consumer experience; Loyalty; Virtual business.

October 22, 2026

Psychosocial Risks and Mental Health at Work in a Banking Institution

Mental health in the workplace is a topic of growing relevance, impacting professional development and employee well-being. A study was conducted to identify aspects and situations that can influence absenteeism levels in a banking institution. The research involved 108 participants and used a structured questionnaire, based on the short version of COPSOQ II by Gonçalves (2021), which contained 45 closed Likert scale questions and two open-ended questions. Data were collected via an online platform and analyzed with a descriptive and quantitative approach. The results revealed a scenario that indicated the need for adjustments in the distribution of activities and in managers’ perception of behaviors and signs of disorders. Important evidence was identified in the participants’ open-ended responses. It was concluded that the study contributed to promoting the perception of a safe environment for mental health in the organization.

Keywords: Absences due to disorders; Professional development; Operational leadership; Medical leave; Mental health.

October 22, 2026

Digital Observatory of the Amazon: Data Science Applied to the Analysis of Social, Economic, Environmental, and Health Indicators

The state of Amazonas faced historical challenges in consolidating the Unified Health System (SUS), stemming from geographical barriers, socioeconomic inequalities, and structural limitations. This study described the development and implementation of a health information geoportal, designed to support monitoring, evaluation, and planning in the Amazonian context. A descriptive case study approach was adopted, based on documentary analysis and secondary data, covering the mapping of functionalities, technical infrastructure, and validation steps of the tool. The Portal was developed with open technologies and integrated multiple national databases, offering interactive dashboards, thematic maps, and a digital document repository. Among the results, highlighted were the georeferenced visualization of key indicators, the possibility of temporal comparative analyses, and the use of the platform in reviewing health plans and redirecting strategic actions. The portal can represent an advance in digital governance and evidence-based management, with potential for replication in other highly complex territories, provided it is accompanied by investments in technological infrastructure, professional training, and inter-institutional coordination.

Keywords: Amazon; Health Evaluation; Digital Health; Health Information Systems.

October 22, 2026

Continuous Learning as a Factor of Organizational Growth in the Perception of Collaborators

The impact of continuous learning and knowledge management on organizational growth was studied in a context of globalization and constant changes. The study’s objective was to analyze employees’ perception of learning culture and its importance for personal and professional development. Data were obtained through an anonymous survey, with a quantitative approach, using a form with “survey” type questions applied to 87 professionals from the technology, health, education, commerce, and services sectors. The results indicated that employees had a positive perception of the contribution of continuous learning to personal and organizational growth, recognizing its benefits for the development of new skills, innovation, and competitiveness. However, it was observed that the incentive for internal knowledge exchange and the offering of practical training still presented moderate levels of agreement. It was concluded that continuous learning significantly contributes to individual and organizational development, suggesting that companies should develop strategic policies for learning focused on new skills, aiming for greater competitive advantage and innovation.

Keywords: Continuous learning; Self-development; Organizational growth; Knowledge Management; Innovation.

October 22, 2026

Statistical Regression in Microgrids for Operational Performance Estimation: Campusgrid Case

A multiple statistical regression model was proposed to estimate the operational performance of UNICAMP’s CampusGrid microgrid, based on internal electrical quantities. The microgrid integrates photovoltaic generation, batteries, and a gas generator. Energy management system data were collected at five-minute intervals over one month. The Operational Reliability Index, defined as the ratio between supplied and demanded energy, was established as the dependent variable. The model was adjusted for the islanded operation mode, evaluating the impact of demand, state of charge, solar generation, and supplied energy. The results revealed high explanatory power, indicating that operational reliability depends on the interaction between supply and demand. It was observed that the nighttime period concentrated the highest rate of partial supply, attributed to the reduction in solar generation. It was concluded that multiple regression offers an interpretable and useful framework for microgrid monitoring and control strategy support.

Keywords: Battery storage; Photovoltaic generation; Reliability Index; Islanded operation; Energy Management System.

October 22, 2026

Forecast of Aggregate Default in the Brazilian Financial System by Sarimax Model with Macroeconomic Variables

Aggregate non-payment in the Brazilian financial system is a crucial indicator of credit risk, the prediction of which is regulated by CMN Resolutions n.° 4.557/2017 and 4.966/2021 for the prospective measurement of expected losses. Aggregate non-payment was modeled and predicted using the SARIMAX model, projecting its trajectory until December 2026 under optimistic, base, and pessimistic scenarios. 164 monthly observations (July/2012 to February/2026) and macroeconomic variables such as unemployment, Selic, IPCA, IBC-Br, and household debt were used. The specification with contemporaneous variables was selected based on information criteria and coefficient stability, and residual diagnostics confirmed statistical adequacy, with a MAPE of 3.12%. Historical non-payment ranged between 4.33% and 7.64%. The estimation identified unemployment (β = 0.2534; p < 0.001) and Selic (β = 0.2128; p < 0.001) as the main positive determinants, with strong temporal persistence (AR(1) = 0.8965). Selic showed increasing correlation with non-payment (r = 0.62 at lag 12), while unemployment demonstrated a contemporaneous effect (r = 0.33). Projections indicated non-payment between 6.30% and 7.98% in December 2026, with a range of 1.68 percentage points, highlighting direct implications for provisioning in financial institutions.

Keywords: Unemployment; Predictive modeling; Monetary policy; Loss provisioning; Time series.

October 22, 2026

Stakeholder Engagement Factors in International Technology Transfer to a Public Hospital

This study described the critical factors perceived for stakeholder engagement in international technology transfer in a public hospital, with a focus on planning. The objective was to describe these conditioning factors, according to the perception of professionals. A single case study was conducted, applying a questionnaire to 25 professionals and using content analysis, complemented by descriptive tabulations. The main obstacles identified were resistance to change (72%), funding (64%), lack of communication (60%), and bureaucracy (40%). Only 28% of respondents recognized the existence of adequate tools for stakeholder management, while 48% denied it and 24% did not know. Exploratory cross-tabulations indicated variations in the assessment of influence according to time of practice, but did not establish causal relationships. Lack of engagement was associated with delays, low adherence, waste of resources, communication failures, compromised knowledge absorption, and risks to patient safety. The conclusions were limited to the perceptions of the researched group and supported preliminary recommendations for intersectoral communication and coordination.

Keywords: Case study; Project Management; Health innovation; Stakeholders; Public sector.

Economy

October 19, 2026

End of the 6×1 scale: what do the numbers allow us to conclude?

Following the promulgation of the end of the 6×1 shift system, the debate on the shorter workday moves to implementation and requires distinguishing social benefits from economic costs. The end of the 6×1 shift system transitioned from legislative discussion to a constitutional change in Brazil. On October 16, 2026, the National Congress promulgated Constitutional Amendment No. 141/2026, originating from Proposal […]

Artificial Intelligence

October 16, 2026

Why study if AI exists?

When answers become easy, learning comes to mean more than just arriving at them. Why do I need to learn this if I can ask an artificial intelligence? This type of question has accompanied education for decades. Why learn an operation if a calculator exists? Why memorize information if it can be found […]

Tax Management

October 15, 2026

Pink tax and the gender of price

How different marketing treatments for men and women are incorporated by consumption taxation Imagine a shelf on which two razors of the same brand and intended for the same purpose are displayed side by side: one has dark packaging and promises strength, precision, and efficiency; the other comes in shades of pink or […]

Health

October 13, 2026

The missing link in health transformation

Why is innovation not enough to drive changes in organizations and health systems Digital transformation has become a strategic priority for health systems worldwide. Electronic health records, artificial intelligence, telemedicine, remote monitoring, and data analysis are now part of the strategic agenda of governments and health organizations. Paradoxically, […]

Marketing

October 09, 2026

Bitcoin or pets: what Artificial Intelligence would not see

A real-life consumer intelligence case shows what gets left out of the brief A few years ago, an American insurance company hired a specialized consumer intelligence firm to answer three questions: The insurer was also considering launching paid insurance in bitcoin. At the time, the consultancy’s workflow was as follows: the analyst created a query […]

School Management

October 06, 2026

The school is not a casino: when trading cards become currency in the playground

Why, for certain students, does school manage to offer so few experiences of achievement? The World Cup is over. Brazil left the competition, the school holidays came, the final was played and the football calendar moved on. It would be reasonable to imagine that, with the end of the tournament and the return to classes, […] would also begin

Energy Transition

September 30, 2026

Renewable energies: the Brazilian advantage

Brazil needs to invest in the capacity to transform natural resources into innovation, productivity, and development. Talking about energy is talking about development. Behind industry, transportation, services, cities, and increasingly artificial intelligence, there is an energy infrastructure that sustains the economy. Understanding a country’s energy matrix is understanding […]

Data Analysis

September 30, 2026

When the data changes, has reality changed?

An analysis of almost 20 million teaching positions-year between 2020 and 2025 shows how changes in population composition and record production can alter the interpretation of historical series Between 2020 and 2025, the number of formal teaching employment positions identified in RAIS (Annual Relation of Social Information), from the Ministry of Labor, […]

Artificial Intelligence

September 28, 2026

Artificial intelligence and the trends

Why do so many people join in on the fun with ChatGPT? I was never the trend person. I still am not. Some pass through my timeline, I see them, I find them interesting but not to the point of trying them out. This was the case, inclusive, with some recent waves of images made by artificial intelligence, like the 80s trend, the one that […]

Tax Management

September 01, 2026

Tax reform and estate planning in the countryside

Efficiency in managing taxes facilitates governance and allows for the organization of family succession. Rural wealth organization has become a strategic measure in light of the tax reform in Brazil, which came into effect this year. The planning of activities and rural wealth management, through tools such as the transition from an Individual to […]

Renata de Gaspari Valdejão Almeida

Artificial Intelligence

August 05, 2026

Inside the company and off the radar: the risks of Shadow AI

According to Microsoft, 75% of professionals already use artificial intelligence at work without company approval. The use of generative artificial intelligence is already part of the corporate routine. Tools like ChatGPT, Copilot, and other platforms help professionals produce texts, analyze data, create presentations, and accelerate tasks in practically all sectors. But this advance […]

Renata de Gaspari Valdejão Almeida

Strategy

July 02, 2026

The future as a competence

Academic interest reinforces the projection of scenarios as an essential tool for decision-making. While some companies are still trying to get the next step right, others are already working with multiple possible futures — and preparing for all of them. But after all, what does it mean to “think about the future”? Is it a trend, a strategy, or a new competence that […]

Renata de Gaspari Valdejão Almeida

Innovation

Technology

May 21, 2026

Agro SP Corridor: innovation, institutional articulation and future

Pecege commands the executive operation of the 400 km ecosystem activities that became known as the Brazilian agro Silicon Valley In 2025, in addition to taking over the management of the Piracicaba Technology Park, the Pecege Institute was also designated to command the executive operation of the State Agribusiness Innovation Corridor activities, the so-called […]

Education

April 27, 2026

How to turn a TCC into a business

New management of the Piracicaba Technology Park offers consulting to MBA graduates The new project director of the Piracicaba Technology Park (PTP), Alexandre Barreto, often says that one of the greatest rewards he has ever had in life was seeing his MBA USP/Esalq students structure companies from their final projects […]

Technology

March 20, 2026

The human repertoire in the age of AI

Critical thinking and creativity are irreplaceable in the work of a marketing professional Artificial Intelligence (AI) is capable of incredible feats, but it does not yet possess the judgment, empathy, intuition, and strategic creativity necessary to do the job of a good marketing professional. In this area, it is important to know how to use the tool to automate […]

Renata de Gaspari Valdejão Almeida

Agribusiness

Innovation

February 23, 2026

ESG as a growth strategy in a cooperative in RS

Santa Clara balances social and environmental responsibility with innovation Founded in 1911 by Italian immigrants in Carlos Barbosa, in the Serra Gaúcha region, Cooperativa Santa Clara is an example that it is possible to balance economic growth with social and environmental responsibility. From a small cheese production company, the organization has evolved into a […]

Renata de Gaspari Valdejão Almeida

Agribusiness

Innovation

January 22, 2026

Brazil free of Foot-and-Mouth Disease: what changes for Brazilian beef?

With a herd of over 200 million head of cattle, the country is currently the leader in exports. Brazil recently received, for the first time, the certification of country free of foot -and-mouth disease without vaccination, conferred by the World Organisation for Animal Health (WOAH). The seal strengthens the country’s position as a reliable supplier of meat for the […]

Renata de Gaspari Valdejão Almeida

Innovation

December 09, 2025

Innovation: the new cycle of the Piracicaba Technology Park

Instituto Pecege assumes management, betting on the creation of new businesses Starting in 2026, Instituto Pecege will assume the management of the Piracicaba Technology Park (PTP). Among its responsibilities will be the administration, operation, maintenance, and expansion of the park, in addition to fostering activities in innovation, entrepreneurship, and science and technology. The choice was made […]

Renata de Gaspari Valdejão Almeida

July 06, 2026

Large-Scale Architecture: DSR-SCRUM Change Management

Mônica Alessandra Guerios; Maria do Carmo Assis Todorov Abstract generated by ResumeAI, an artificial intelligence solution developed by Instituto Pecege focused on synthesis and writing. Architecture and urbanism are configured as essential bases in an increasingly complex and challenging urban space, where the intrinsic connection between social, environmental, and technological issues […]