Predictive Model for Identifying Leaks in Water Distribution Networks via Pressure Monitoring

doi:10.22167/2675-6528-202602654

Article

September 30, 2026

Predictive Model for Identifying Leaks in Water Distribution Networks via Pressure Monitoring

Predictive Model for Identifying Leaks in Water Distribution Networks Via Pressure Monitoring

Ana Carolina Porto; Elisa Morandé Sales

DOI: 10.22167/2675-6528-202602654

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

Real water losses in distribution networks represent a crucial challenge for the operational efficiency of water utilities, especially in light of the reduction targets established by Law No. 14.026/2020. A predictive model was developed to identify leaks in water distribution networks based on pressure data monitoring, with the objective of estimating the magnitude and the most probable location of the events. The methodology integrated hydraulic simulations, operational data, and deep learning techniques, using a hybrid CNN-LSTM architecture. The model was developed and evaluated on the virtual distribution network L-Town, from the BattLeDIM (2020) competition, combining synthetic data generated via the WNTR library with pressure time series for external evaluation. The results demonstrated the viability of the approach, with an R² of 0.7947 and a zone accuracy of 70.05% in the synthetic domain, and a Pearson correlation of 0.7449 in the external temporal macro-trend evaluation. The domain divergence observed, with a drop in localization accuracy from 70.05% to 18.02% in the external evaluation, delimited the current scope of the model and indicated the need for improvements for its implementation in operational networks. The model presents itself as a promising tool for loss management, standing out in leak quantification, but requires improvements for exact field localization.

Keywords: Leak detection; Machine learning; Real water losses; Water supply system.

1. Introduction

Due to the growing concern associated with climate change and water scarcity scenarios, the conservation of natural resources and the conscious use of water have been gaining prominence nationally and internationally. In this context, with the update promoted by Law nº 14.026/2020, the new legal framework for basic sanitation came into effect in Brazil, establishing guidelines for regulation, universal access, and efficient service delivery, including the reduction of water loss rates. In addition, in 2025, Reference Standard nº 15/2025 from the National Water and Basic Sanitation Agency (ANA) was approved, which defines specific guidelines for the management, progressive reduction, and control of losses in supply systems.

By definition, in an urban supply system, water losses are accounted for through the difference between the total volume inserted into the system — corresponding to produced and imported water — and authorized consumption, being classified into real losses and apparent losses. Real loss, also known as physical loss, corresponds to the portion of water that does not reach the final consumer, being mainly associated with leaks in pressurized elements of the system, such as distribution networks and service lines. In turn, apparent loss, or commercial loss, refers to the volume of water effectively consumed, but not properly measured, accounted for, or authorized, being related to inaccuracies in metering, data handling errors, and unauthorized consumption. In addition to these categories, the concept of “unbilled water” stands out, which represents the total water not converted into billing, encompassing both real and apparent losses as well as authorized unbilled consumption (Pearson, 2019).

The study conducted by Liemberger and Wyatt (2018) indicates that the world volume of unbilled water was, in 2019, on the order of 346 million cubic meters per day. According to the authors, besides representing a financial concern, the high rate of losses in the current context of growing scarcity and climate change compromises the ability of supply concessionaires to achieve their universalization goals, as well as to deliver a reliable, quality, and affordable service.

Nationally, the sanitation ranking published by the Trata Brasil Institute (2025) analyzed the indicators of the 100 most populous municipalities with reference values from SNIS/SINISA 2023. The report presented an average index of distribution losses of 45%, which represents a significant worsening compared to the 35% computed in the previous year, also higher than the national average disclosed in SINISA (2023), which was 40%.

Regarding real losses specifically, leaks in distribution networks and building connections are considered the most representative portions, and can be classified as visible, non-visible, or inherent. While inherent leaks are conditioned by the physical and operational situation of existing infrastructures, visible and non-visible leaks are directly linked to the Average Service Time (AST). Thus, the rapid detection of new events is an important strategy in reducing the volume of lost water, which varies according to the time between the awareness of a new rupture until its effective location and repair (Manzi, 2017).

Currently, several methods are available for active leak detection, such as noise correlation technologies, in-line techniques (cabled or free-standing systems), electronic amplified listening devices, drones, ground-penetrating radar (GPR), infrared/thermal cameras, satellite imagery, optimization tools (hydraulic modeling), and software-based solutions (Hamilton and Charalambous, 2020). However, despite the variety of technologies, good performance often requires more assertive guidance. Due to high complexity, lack of precise diagnostics, and low information availability, the available technologies do not always deliver practical, efficient, and effective results.

It is known that the flow rate of leaks in a given pipeline increases as a function of the increase in fluid pressure. This occurs because the leak area is not fixed, oscillating according to the pressure variation to which the pipeline is subjected, also influenced by the specificities of the network material (Van Zyl et al., 2017). Although this correlation is known, the applicability of tools that enable a precise diagnosis of this impact in real systems is not trivial. When analyzing pressure data, several variables make it difficult to separate the “signatures” of each recorded event, which, in addition to leaks, include variations in consumption and operational events, such as filling reservoirs, activating pumps, opening and closing valves, among others.

Misiunas (2005), in a diagnosis that remains valid, observed that despite the efforts made to modernize the operation of water supply systems, this is still one of the areas that most requires development within urban infrastructure. In recent years, technological advances have significantly driven the development of tools to assist in the detection of leaks in water distribution networks, such as the use of artificial intelligence and deep learning techniques.

The objective of this work is to structure a predictive model capable of identifying leaks in water distribution networks, both in terms of magnitude, expressed by the emitter coefficient, and location, represented by the delimitation of a more probable area for the occurrence of the event. The proposed approach seeks not only to improve the diagnosis of real loss volume in supply systems, but also to increase the precision and speed in the detection and localization of leaks. By prioritizing and enhancing the use of the measurement infrastructure already installed in existing systems, the work aims to develop a tool to support operational efficiency, loss reduction, and improved intelligent network management.

2. Material and Methods

The methodology adopted in this work was characterized by a quantitative and predictive approach, with the objective of developing an artificial intelligence model for the detection and localization of leaks in water distribution networks. The study integrated computational hydraulic modeling for the generation of synthetic data, time series preprocessing, and the development of a hybrid deep neural network architecture.

The empirical object of the research was the virtual distribution network L-Town, made available by the international scientific competition BattLeDIM (2020). This fictitious network consists of 782 demand nodes, 1 reservoir, and 2 supply sources. The network has an extension of 43.16 km, modeled in 905 sections, and includes 3 valves and 1 pump. For system monitoring, pressure data from 33 sensors distributed throughout the network were used.

Due to the scarcity of real labeled data from leaks in operational environments, the generation of a synthetic dataset was chosen. 10,000 leak scenarios were simulated, each lasting 24 hours with a hydraulic and reporting time step configured for 900 seconds (15 minutes), resulting in time series of 96 points per sensor. The simulations were performed using the “Pressure-Driven Demand” (PDD) hydraulic calculation engine.

The magnitude of the leaks in the synthetic dataset was defined by the emission coefficient, randomly varying between 0.0001 (small leak) and 0.0015 (medium leak). Additionally, Gaussian noise (mean 0, standard deviation 0.2) was added to the simulated pressure readings, with the aim of representing uncertainties and measurement errors of real sensors. For each scenario, only one active leak was simulated per node.

The spatial discretization of the localization problem was carried out by grouping the network nodes into geographical zones. The K-Means algorithm was applied to the Cartesian coordinates (X,Y) of the nodes, defining 10 clusters (zones). This approach resulted in approximately 4.3 km of network per sector, transforming the exact localization problem, of high complexity, into a multiclass classification problem by zones.

Before feeding into the neural network, the data underwent three treatment steps. Firstly, residual calculation was performed by subtracting a “baseline” corresponding to the network’s behavior without leaks from the simulated (leaked) pressure time series, thus isolating the hydraulic signature of the anomaly.

Next, the “MinMax Scaler” technique was applied for normalization, rescaling the residual values to an interval that facilitated gradient convergence during training. Finally, for validation with real data from 2019 (external evaluation), a sliding window technique with a temporal step size of 96 was used, ensuring compatibility with the input format of the trained model.

The model was built from a hybrid proposal, integrating Convolutional Neural Networks (CNN-1D) and Recurrent Neural Networks (LSTM). The proposed architecture consisted of a “Multi-Task Learning” structure, with the objective of simultaneously exploring local temporal patterns and long-term dynamic dependencies in the pressure signals.

The model input consisted of a tensor of dimension (96, 33), representing 96 time instances (24 hours of measurements recorded every 15 minutes) from 33 pressure sensors. The network structure was composed of sequential layers, starting with a 1D convolutional layer with 64 filters and kernel size equal to 3, followed by a ReLU activation function and MaxPooling with a factor of 2, for spatial feature extraction.

Subsequently, a Long Short-Term Memory (LSTM) layer with 50 units processed the features extracted by the CNN-1D, capturing long-range temporal dependencies and the daily dynamics of the pressures. An intermediate dense layer of 64 neurons connected to two independent output layers: one for leak zone classification (Output_Localizacao, with 10 neurons and Softmax activation) and another for leak intensity regression (Output_Magnitude, with 1 neuron and linear activation).

All development was carried out in the “Google Colab” environment, using the Python language (version 3.12.13). The “Water Network Tool for Resilience” (WNTR, v1.4.0) libraries were used for hydraulic simulation, and “TensorFlow” (v2.21.0) for building predictive models. The synthetic dataset was divided into 80% for training and validation, and 20% for testing.

The model was compiled using the Adam optimizer. The loss function was weighted, assigning a weight of 3.0 to the classification loss (“sparse_categorical_crossentropy”) and a weight of 1.0 to the regression loss (“mean_squared_error”), prioritizing the localization task. Training occurred over 100 epochs, with a batch size of 64 samples.

To quantify the model’s performance, specific metrics were adopted for the regression (magnitude) and classification (localization) tasks. For regression, the Coefficient of Determination (R²), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) were used. For classification, Zone Accuracy (Exact Match), Area Under the ROC Curve (AUC-ROC), Precision, Recall (Sensitivity), F1-Score, and the Confusion Matrix were employed.

The final validation of the model trained exclusively with synthetic data was performed by applying it to the real monitoring data from the year 2019 (2019_SCADA_Pressures.csv) of the L-Town network. Its generalization capacity was evaluated through comparison with the ground truth of real leaks that occurred in that year (2019_Leakages.csv), including Temporal Correlation for the macro-trend of leaks.

3. Results and Discussion

The implementation of the hybrid predictive model, which integrates Convolutional Neural Networks (CNN) and Recurrent Neural Networks (LSTM), was carried out and evaluated in two distinct stages for the detection and localization of leaks in water distribution networks. Firstly, a quantitative validation was performed using a reserved test set, corresponding to 20% of the synthetic dataset of 10,000 generated scenarios. Subsequently, an external validation was conducted, employing real monitoring data from the L-Town network for the year 2019. The model training extended for 100 epochs, with the objective of simultaneously executing the classification of the leak occurrence zone and the regression of the emission coefficient, which represents the magnitude of the event.

Performance in the Training Process

The model training demonstrated robust and stable convergence over 100 epochs. The accuracy for the localization task, which was initially below 20% in the early epochs, showed significant evolution, stabilizing at approximately 70% in validation at the end of the process. This progressive improvement indicates the model’s ability to learn and refine its localization predictions. In parallel, the total loss function, or “loss”, exhibited a consistent decay, starting from values close to 7.0 and reaching approximately 2.0. The proximity between the training and validation curves for loss and accuracy suggests that the model did not incur severe overfitting, indicating good generalization of the learned patterns to unseen data during training, as illustrated in graphical analyses of accuracy and total loss.

Internal Validation (Synthetic Dataset)

The internal validation, conducted on the synthetic dataset, allowed for a detailed evaluation of the model’s performance in a controlled environment. In the task of estimating leak magnitude, the model demonstrated satisfactory performance. The Coefficient of Determination (R²) reached 0.7947, meaning the model was able to explain approximately 80% of the data variance in the synthetic environment. This result is crucial, as it indicates that the hybrid CNN-LSTM architecture can effectively capture the relationships between pressure variations and leak intensity, as expected for a robust predictive model.

Additionally, the error metrics for magnitude estimation reinforced the model’s accuracy. The Mean Absolute Error (MAE) was 0.0987, and the Root Mean Squared Error (RMSE) was 0.1320. These relatively low values indicate that the model’s predictions for leak magnitude are, on average, close to the actual values. The graphical analysis of the scatter plot between predicted and actual values confirmed this trend, with most points clustering near the ideal line, although a slight dispersion was observed at intermediate magnitudes, suggesting that the model may perform slightly less accurately for medium-sized leaks.

The localization task, which involved classifying leaks into one of the 10 predefined geographic zones of the network, presented a greater challenge, but the model achieved a Global Accuracy of 70.05% on the synthetic test set. This value is mathematically convergent with the weighted “Recall”, indicating a good ability to correctly identify most leaks within their respective zones. The AUC-ROC metric (Area Under the Receiver Operating Characteristic Curve) reached 0.9690 (Macro Average), which suggests a high capacity of the model to distinguish between the different zone classes, classifying the correct zone with high probability.

Other weighted classification metrics were also evaluated, providing a comprehensive view of the model’s performance in the localization task. The weighted Precision was 0.7268, the weighted Recall (Sensitivity) was 0.7005, and the weighted F1-Score reached 0.7040. These values indicate a good balance between the model’s ability to make correct predictions (Precision) and to identify all positive cases (Recall), with the F1-Score being a useful harmonic mean for evaluating performance in scenarios with imbalanced classes, which is common in anomaly detection problems.

The analysis of the normalized confusion matrix revealed important patterns in classification errors. It was observed that the errors are not random, but tend to concentrate in adjacent zones or in regions with similar hydraulic characteristics. This suggests that, although the model may not identify the exact zone, it frequently points to an area geographically or hydraulically close to the actual leak. The average spatial distance error, calculated at 168.02 meters in the test set, corroborates this observation, indicating that, even in cases of incorrect classification, the model’s prediction remains relatively close to the true location, which can be useful for directing field teams.

External Validation (Real Scenario – 2019 Data)

The application of the model to the real pressure monitoring data from 2019, originating from the 33 SCADA sensors of the L-Town network, followed the same preprocessing pipeline used in training. This process included the subtraction of a simulated hydraulic baseline, data normalization by a “scaler” fitted to the synthetic data, and the formatting of the 3D Tensor with 96 time steps, corresponding to 24 hours of measurements. The model’s magnitude output estimated the normalized emitter coefficient C, as defined by EPANET’s leakage equation, which relates leakage flow rate to pressure and the emitter coefficient.

The comparison of the magnitude of leaks in the real scenario was performed in the trend domain, using a moving average. This approach was necessary because the 2019 benchmark recorded flow rates in cubic meters per hour, while the model estimated the normalized emission coefficient. Without knowledge of the instantaneous local pressure, the absolute values were not directly comparable. Both curves, the real and the estimated one, were normalized and smoothed by a 24-hour moving average, allowing for an evaluation of the model’s ability to capture the periods of increase and decrease in leaks throughout the year. The Pearson correlation obtained was 0.7449, indicating that the temporal macro-trend of real leaks was satisfactorily captured by the model, which is a promising result for loss management.

However, spatial validation in the real-world scenario evidenced considerable limitations compared to performance in the synthetic environment. Zone accuracy, which measures the exact correspondence between the predicted zone and the actual zone of the pipe with the highest leakage, was only 18.02%. This value represents a significant drop compared to the 70.05% obtained in the synthetic dataset. The average localization distance error also increased considerably, reaching 807.98 meters. Despite this reduction in precision, the model demonstrated some capacity to identify higher severity events, correctly identifying the zone of the pipe with the highest recorded leakage volume in 2019.

The consolidated metrics from the validation with the real scenario highlight the performance divergence between the synthetic and real environments. The temporal correlation (macro-trend) remained at 0.7449, confirming the model’s ability to track magnitude variations over time. However, the zone accuracy (Exact Match) of 18.02% and the mean distance error of 807.98 meters for localization indicate that the transfer of learning from the synthetic to the real environment still faces significant challenges, particularly regarding the spatial precision of leaks. This discrepancy between the accuracy of synthetic data (70%) and real data (18%) points to model weaknesses that need to be addressed in future work.

The fragilities identified in the model, which resulted in the discrepancy between performance on synthetic and real data, were listed as crucial points for future improvements. Firstly, the model was trained with the premise of a single active leak per scenario in the synthetic dataset. However, real data from 2019 frequently contain multiple simultaneous leaks, which can overlap pressure signatures and confuse the zone classifier. The incorporation of scenarios with multiple simultaneous leaks in the structuring of synthetic data is, therefore, a necessity to increase the model robustness in operational environments.

Another point for improvement lies in the aggregation of seasonality to the simulation. The “baseline_series” was generated by a single static simulation, while the real data presents seasonal consumption variations that the static baseline cannot capture. The lack of seasonality representation in training can lead to false positives or reduced sensitivity in leak detection across different periods of the year. Furthermore, the insertion of signatures from different noises, such as valve maneuvers and reservoir filling operations, is fundamental. The “scaler_X” was defined on synthetic data with controlled Gaussian noise (σ=0.2), but the real data has a different and unknown noise distribution, which affects the model’s ability to generalize.

The division into 10 zones by geometric K-Means also proved inadequate, as it does not account for the real hydraulic topology of the network. Consequently, nodes belonging to the same geometric zone may be hydraulically distant, which compromises localization accuracy. To complement these improvements and add traceability to topological variations without excessively increasing computational cost, some recommendations were proposed. It is suggested to replace statistical zone generation (K-Means) with a topological analysis, grouping nodes not by straight-line geometric distance, but by the length of the network’s piping and physical barriers, such as valves, pumps, and capping.

Additionally, the integration of “Physics-Guided Machine Learning” (PGML) is recommended, injecting the Pressure Differential (ΔP) between adjacent sensors directly into the pre-processing step. This approach aims to provide notions of topology and flow direction to the current architecture, seeking to capture spatial generalization benefits at a lower computational cost than more complex approaches. Another proposal is the Post-Prediction Hydraulic Triangulation, which would use artificial intelligence strictly for macro-zone and flow rate (magnitude) detection, coupling a deterministic algorithm at the model’s output that crosses these predictions with empirical equations, such as Hazen-Williams, to locate the exact node analytically.

Finally, the creation of an Off-line Sensitivity Matrix was recommended. This matrix would previously map the directional range of leaks during typical nighttime minimum flow scenarios, between 1 AM and 5 AM. Such a matrix would act as a filter for noise elimination and analytical restriction of false positives during field operation, optimizing the response and accuracy of interventions. These recommendations aim to enhance the model’s ability to handle the complexity of real networks and the variability of operational data, bringing it closer to practical and efficient application in loss management.

In summary, the predictive model developed, based on a hybrid CNN-LSTM architecture, proved to be a promising tool for managing losses in water distribution networks. Its effectiveness was verified in estimating the magnitude of leaks, with a Coefficient of Determination (R²) of 0.7947 on synthetic test data and a strong correlation of 0.7449 in the temporal macro-trend of external validation data. Although it showed good generalization capacity for localization in a controlled environment (AUC-ROC of 0.9690 and mean error of 168 meters), its application in a real-world scenario revealed limitations in spatial accuracy, with a mean localization error of 807.98 meters, indicating the need for improvements for its field deployment as an exact localization tool.

4. Conclusion

This study sought to structure a predictive model to identify leaks in water distribution networks, estimating their magnitude and delimiting the most probable location based on pressure data monitoring. It was found that the developed hybrid CNN-LSTM architecture demonstrated effectiveness in quantifying leak magnitude, achieving a Coefficient of Determination (R²) of 0.7947 on synthetic data and a Pearson correlation of 0.7449 in the temporal macrotrend in a real-world scenario. For localization, the model obtained a zone accuracy of 70.05% in the synthetic environment, with an average distance error of 168.02 meters, indicating its generalization capability under controlled conditions. The main contribution lies in offering a promising tool for loss management, capable of improving the diagnosis of real loss volume and supporting operational efficiency by prioritizing the use of existing measurement infrastructure.

However, the application of the model to real monitoring data revealed significant limitations in spatial precision, with zone accuracy dropping to 18.02% and the average localization distance error increasing to 807.98 meters. This domain divergence was attributed to the single leak premise in synthetic training, the absence of seasonality in the simulation, and the inadequacy of K-Means geometric sectorization for the real hydraulic topology. To enhance the model’s robustness and applicability in operational environments, it is recommended to incorporate scenarios with multiple simultaneous leaks in the synthetic data, evolve the sectorization through topological analysis, and integrate Physics-Guided Machine Learning (PGML) techniques to provide notions of topology and flow direction. Hydraulic Post-Prediction Triangulation and the creation of an Off-line Sensitivity Matrix are also suggested to refine exact localization and filter noise, bringing the model closer to an accurate field localization tool.

Bibliographic References

BattLeDIM, 2020 [Referência completa não encontrada no documento original]

Brasil. Agência Nacional de Águas e Saneamento Básico. Norma de Referência nº 15, de 2025. Estabelece diretrizes para a regulação dos serviços de saneamento básico, com foco na redução e controle de perdas de água. Brasília, 2025.

Hamilton, S.; Charalambous, B. 2020. Leak Detection: Technology and Implementation. 2ed. IWA Publishing, London, UK.

Instituto Trata Brasil; GO Associados. 2025. Ranking do Saneamento do Instituto Trata Brasil de 2025 (SNIS/SINISA 2023).

Liemberger, R.; Wyatt, A. 2018. Quantifying the global non-revenue water problem. Water Supply 18(4): 1184-1193.

Manzi, D. 2017. Detecção e Localização de Rompimentos em Redes de Distribuição de Água a partir de Análise dos Sinais de Pressão em Regime Transitório. Tese de Doutorado em Engenharia Civil. Universidade Estadual de Campinas, Campinas, SP, Brasil.

Misiunas, D. 2005. Failure Monitoring and Asset Condition Assessment in Water Supply Systems. Tese de Doutorado em Automação Industrial. Lund University, Lund, Suécia.

Pearson, D. 2019. Standard Definitions for Water Losses. IWA Publishing, London, UK.

SINISA, 2023 [Referência completa não encontrada no documento original]

SNIS/SINISA, 2023 [Referência completa não encontrada no documento original]

Van Zyl, J. E.; Lambert, A. O.; Collins, R. 2017. Realistic modeling of leakage and intrusion flows through leak openings in pipes. Journal of Hydraulic Engineering 143(9): 04017030-1-04017030-7.

Article originating from the Final Course Work 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 […]