Between sensors and algorithms: the new logic of industrial maintenance

Column

Data Science

Data Science Analytics

Technology

June 09, 2025

Between sensors and algorithms: the new logic of industrial maintenance

Previously restricted to highly technological environments, predictive support now presents itself as a concrete and accessible possibility

Predictive maintenance has been gaining prominence in modern industry, by allowing the anticipation of failures and the optimization of the availability of productive assets. Unlike traditional models, such as corrective and preventive maintenance, this approach uses real data from operation and intelligent algorithms to indicate the ideal moment for intervention, reducing unplanned downtime and operational costs (Es-sakali et al., 2022; Poór & Basl, 2019).

With the advancement of Industry 4.0, embedded sensors, industrial networks, and monitoring systems have started to generate large volumes of data in real-time, which has opened space for the application of data science techniques, such as machine learning, neural networks, and statistical models (DalzCochio et al., 2020; Esteban et al., 2022). In this context, data science acts as a means to convert raw data into useful information for maintenance decisions.

The growing availability of operational data and the maturation of analytical tools have transformed predictive maintenance into a concrete application of data science within industrial operations. Models such as recurrent neural networks, support vector machines, classification algorithms, and deep learning techniques are already successfully used to predict failures, estimate the remaining useful life of equipment, and classify degradation patterns (Nikfar et al., 2022; Singh et al., 2023; Wang et al., 2024).

 Furthermore, more recent approaches, such as transfer learning, continual learning, and federated learning, have been explored to overcome limitations like labeled data scarcity, non-stationary operational environments, and privacy constraints in decentralized applications (Ahn et al., 2023; Azari et al., 2023; Hurtado et al., 2023). These advances demonstrate that data science occupies a strategic position in the transition from reaction-based maintenance models to prediction- and optimization-oriented strategies.

Applications

The application of data science in predictive maintenance is already a reality in several industrial sectors, with emphasis on the areas of energy, manufacturing, and transportation, and the aerospace sector. In low-voltage industrial motors, for example, algorithms such as neural networks, random forest and support vector machines have been successfully used to detect failures and classify them with an accuracy greater than 95%, even with relatively small datasets (Nikfar et al., 2022).

In more complex industrial environments, such as automotive production lines and wind turbines, multivariate sensors integrated with deep learning models, like CNNs and LSTMs, have enabled more robust diagnostics and early interventions, with near 100% accuracy in anomaly detection (Abdullahi et al., 2024; Gawde et al., 2024). These results illustrate how adequate data collection and processing, combined with well-calibrated predictive models, can reduce unexpected failures and increase asset availability with a direct impact on productivity.

Beyond conventional predictive models, the use of digital twins has been consolidated as a sophisticated extension of data science applied to maintenance. The term “digital twin” refers to the creation of a virtual replica of a physical asset, such as an engine or turbine, which mirrors its behavior in real-time through the integration of sensors and computational models. Just as a human twin shares genetic characteristics, the digital twin shares operational data with its physical counterpart. By integrating physical and data-driven models, digital twins enable real-time simulations , continuous diagnostics, and accurate estimations of the remaining useful life of critical components, such as engines, gearboxes, and battery systems (Singh et al., 2023; van Dinter et al., 2022; Zhong et al., 2023).

In recent studies, distributed digital twin architectures have been successfully applied to wind turbines, combining physical sensors, cloud computing, and deep learning models for high-accuracy, low-latency fault prediction (Abdullahi et al., 2024). In another example, the conversion of time series into images using techniques such as Gramian Angular Fields, subsequently analyzed by convolutional networks, allowed for the detection of critical faults in industrial motors with 100% accuracy, outperforming traditional models such as SVM (Kiangala & Wang, 2020).

These cases demonstrate that, when well applied, the tools of data science offer not only predictive accuracy, but also visualization, transparency and decision-making that is more agile.

Scale

Given the diversity of equipment, operational contexts, and privacy restrictions, more recent approaches have sought to make predictive models more generalizable, scalable, and collaborative. Federated learning, for example, has been used to train predictive models distributed across multiple factories without the need to centralize data, protecting confidentiality and adapting algorithms to the peculiarities of each industrial unit (Ahn et al., 2023). Transfer learning, on the other hand, has proven useful for leveraging previously trained models on similar equipment, reducing the need for large volumes of labeled data and accelerating adoption in new assets (Azari et al., 2023).

In dynamic scenarios, where the system’s behavior changes over time, continuous learning has been explored to ensure models update themselves without losing acquired knowledge, handling different operational regimes and variations in sensor data (Hurtado et al., 2023). These applications demonstrate the maturity of data science in the field of maintenance, not just as a prediction tool, but as adaptive and strategic technology.

Challenges

Despite the advances, the implementation of predictive maintenance based on data science faces challenges. One of the main obstacles is related to the quality and availability of data. Old equipment is often not prepared to provide real-time data, and even when sensors are available, collection can be affected by noise, losses, and inconsistencies, compromising the effectiveness of predictive models (Dalzochio et al., 2020; Esteban et al., 2022).

Furthermore, the scarcity of failure records is a recurring problem, as most assets operate for long periods without critical failures. This limitation particularly affects supervised models, which rely on labeled data to learn degradation patterns (Azari et al., 2023; Divya et al., 2023).

In response to this scenario, many studies have invested in hybrid or unsupervised approaches, as well as in the use of synthetic data and digital twins as alternatives to simulate failures and enrich training sets (Singh et al., 2023; Zhong et al., 2023).

Another challenge concerns the integration of predictive models into the culture and decision-making processes of organizations. Even when data is available and analytical capacity exists, it is common to find resistance from managers and operational teams to fully trust the recommendations provided by artificial intelligence algorithms, especially when these models function as black boxes — although the algorithms generate predictions or diagnoses, they do not transparently explain how they reached those conclusions, which hinders the validation by professionals who rely on this information to make critical decisions (Chen et al., 2021; Gawde et al., 2024).

To mitigate this problem, several approaches have incorporated explainable artificial intelligence (XAI) techniques, such as LIME and variable importance analysis, allowing experts to understand the factors that led to the prediction of a failure or the triggering of an intervention (Ahn et al., 2023; Gawde et al., 2024). Furthermore, the adoption of predictive technologies often encounters the absence of qualified personnel, requiring new professional profiles with skills in data science, maintenance, and digital technologies (Poór & Basl, 2019). This reinforces the need for training programs and cultural change that align digital transformation with strategic objectives of reliability and operational performance.

Additionally, regulatory and financial aspects also impose barriers to the large-scale adoption of predictive maintenance based on Data Science. In sectors such as railway, aeronautics, and energy, the implementation of predictive systems must meet rigorous safety and traceability requirements, which demands technical validation of models and transparency in decision-making (Rokhforoz & Fink, 2021; Scott et al., 2022).

From the economic point of view, the initial investment required for the acquisition of sensors, network infrastructure, cloud storage, and analytical model development can be high, especially for small and medium-sized enterprises (Chen et al., 2021; Meng et al., 2022). Although several studies demonstrate a positive return on investment over time, the absence of standardization in architectures, platforms, and protocols hinders the replication of solutions among companies and sectors (Ton et al., 2020; van Dinter et al., 2022). In response, recent initiatives have sought to develop generic frameworks and maturity models that assist organizations in planning the gradual transition from traditional to predictive maintenance (Mesarosova et al., 2022; Poór & Basl, 2019).

Conclusion

The consolidation of predictive maintenance as a strategic practice in industrial organizations has been directly driven by the advancement of data science. Statistical models, machine learning algorithms, and artificial intelligence techniques have proven highly effective in predicting failures, estimating the remaining useful life of assets, and generating early alerts based on real operational data (Esteban et al., 2022; Nikfar et al., 2022; Singh et al., 2023).

At the same time, Data Science has enabled the development of more sophisticated solutions, such as digital twins and distributed architectures with edge computing, capable of processing information in real-time and making decisions locally (Abdullahi et al., 2024; Zhong et al., 2023). Predictive maintenance, previously restricted to highly technological environments, now presents itself as a concrete and accessible possibility, provided it is structured based on relevant data, adequate methods, and an integrated organizational vision.

The future of predictive maintenance is directly linked to the evolution of increasingly adaptive, collaborative, and explainable approaches. Trends such as continuous learning, federated learning, and transfer learning are expected to gain prominence, as they allow for constant model updates even in decentralized environments, with little labeling or subject to operational changes (Ahn et al., 2023; Azari et al., 2023; Hurtado et al., 2023).

Furthermore, the integration between data science and areas such as logistics, production, and asset management enables a broader view, connecting technical decisions to business strategies (Meng et al., 2022; Wang et al., 2024). In this context, the ability to interpret models, ensure data quality, and train professionals capable of intermediating technical and analytical knowledge will be decisive for success. More than a technological evolution, predictive maintenance represents a shift in mindset: from reactive repair to intelligent anticipation, from isolated analysis to data-driven decision-making. In this new scenario, data science ceases to be a support tool and begins to occupy a central place in engineering, asset management, and the construction of resilient industrial operations. Organizations that know how to unite technical knowledge, quality data, and strategic vision will be better prepared to compete in increasingly demanding, dynamic, and information-driven industrial environments.

To access the references of this text click here.

Who wrote this column

José Erasmo Silva

José Erasmo Silva é professor, formado em Matemática e Administração, com mais de 25 anos de experiência em gestão empresarial e de pessoas. É mestre e doutor em Administração, com foco em Finanças, e especialista em Data Science e Analytics e em Finanças e Controladoria. Realizou pós-doutorado na Universidade Federal da Bahia (UFBA). Atualmente, atua como professor orientador no MBA em Data Science, Inteligência Artificial e Analytics da USP/Esalq e leciona na EEP/FUMEP e na rede estadual de ensino de São Paulo.

You may also like

November 17, 2026

Detection and Diagnosis of Diseases in Corn and Rice in Mozambique with Deep Learning

Mozambican agriculture depends on corn and rice, fundamental crops for food security, but their productivity is strongly affected by phytosanitary diseases, whose early and accurate identification remains a challenge in resource-limited contexts. The study aimed to develop and compare Deep Learning-based models for the automatic detection and diagnosis of diseases in corn and rice leaves in Mozambique. Three approaches were used: a classic Convolutional Neural Network (CNN) developed from scratch, and two transfer learning models, ResNet50 and MobileNetV2. The dataset consisted of images from public databases and collected in the field, which underwent cleaning, balancing, and artificial augmentation steps. The models were trained and evaluated with standard performance metrics, including accuracy, precision, recall, F1-score, Log Loss, and Cohen’s Kappa. The classic CNN presented the best overall performance, with 94% precision, recall, and F1-score, and 93% accuracy and Cohen’s Kappa. MobileNetV2 achieved 91% F1-score and 90% accuracy, standing out for its lower Log Loss (0.30) and training time (6h32min59s). ResNet50 reached 80% F1-score and 79% accuracy. It was concluded that Deep Learning techniques have high potential to support phytopathological diagnosis in real agricultural contexts, contributing to faster and more accurate decisions in the management of corn and rice diseases.

Keywords: Deep learning; Rice; Plant diseases; Corn; Computer vision.

November 17, 2026

Artificial intelligence in accounting operational routines and financial efficiency

The incorporation of Artificial Intelligence (AI) and automation tools has reconfigured the routines of Brazilian accounting offices, impacting productivity, error mitigation, and financial efficiency. These impacts were analyzed in a small accounting firm, specialized in serving service providers in the medical field in São Caetano do Sul, São Paulo. The research adopted a qualitative approach, with a single case study design and incorporated units of analysis. Data were collected through a structured questionnaire, applied via Google Forms between January and February 2026, to five professionals in the operational structure of the firm in a census survey. The treatment followed content analysis with a priori and emergent categories. All respondents declared the use of automated tools or tools with AI resources. The perception of reduced execution time, decreased operational errors, rework, and better use of the team’s time converged in 80% of affirmative responses and 20% partial responses. The most benefited activities were bank reconciliation, import of fiscal documents, and tax assessment, with visible gains on typing errors, duplicate entries, and incorrect classification. Human intervention remained indispensable in tax planning, scenario analysis, and verification of ancillary obligations. The overall evaluation of AI’s contribution was predominantly favorable, with moderation linked to the complexity of fiscal regulations.

Keywords: Content analysis; Process automation; Accounting 4.0; Small business; Digital transformation.

November 17, 2026

Factors influencing schedule management in sewage treatment plant “retrofit” projects

Projects involving “retrofit” in Sewage Treatment Plants (STP) are characterized by high complexity, operational restrictions, and multiple technical interfaces. The study analyzed the perception of the professionals involved regarding the management and adherence to deadlines in “retrofit” projects for sewage treatment plants, considering technical, operational, and organizational factors. The research was characterized as an applied study, with a quali-quantitative approach, designed as a field survey. Data analysis was performed using descriptive statistics and qualitative categorization, employing SWOT, risk matrix (Probability x Impact), and 5W2H action plan tools. The results indicated a high perception of clarity (77.8%) and realism of deadlines (100% agreement), evidencing consistent initial planning. However, 77.7% of respondents pointed out that rework and scope changes directly impacted deadline adherence. The integration between engineering, manufacturing, and Production Planning and Control (PPC) showed an intermediate evaluation (77.7% partial/total agreement), indicating fragility in interfunctional coordination. The risk matrix highlighted as critical: scope changes, operational limitations of the STP, and technical complexity, all with a high level. As a central improvement, the standardization of scope change control, analysis, and approval was proposed, aiming to reduce rework and increase schedule predictability.

Keywords: Performance; Integration; Planning; Rework; Risk.

November 17, 2026

B2C Acquisition in diagnostic medicine: strategic partnerships with specialized medical offices

The diagnostic medicine sector in Brazil operates with high dependence on health plan providers, which limits the financial autonomy of service providers. This study proposed and analyzed a direct patient acquisition model (B2C) for a highly complex diagnostic medicine institution, structured through partnerships with Oncology and Cardiology practices and operationalized by a personalized customer service called concierge. The investigation adopted a mixed-methods approach in three phases: a field survey with a structured questionnaire applied to 87 patients, mapping of the operational flow using “Service Blueprint”, and analysis of the conversion rate from 312 medical referrals registered in the pilot project. The results indicated that trust in medical referral achieved 92.0% agreement, establishing itself as the main factor for adherence. The convenience of performing the exam on the same day reached 78.2%, and the willingness to pay directly was 55.2%, with polarization associated with the predominance of health plan users. The “Service Blueprint” identified three critical points in the transitions between stages of the patient journey. The overall conversion rate was 51.0%, with superior performance in Oncology (58.2%) compared to Cardiology (45.0%), and the greatest loss was concentrated in the acceptance and scheduling stage (28.3%). It was concluded that the model demonstrated possible operational viability and indications of patient acceptance, conditioned by price transparency and a reduction in the interval between consultation and approach by the concierge.

Keywords: Direct payment; Patient journey; Supplementary health; Concierge service; Conversion rate.

November 17, 2026

The role of the school manager in promoting an inclusive culture: leadership and educational policies

The research analyzed the school manager’s performance in promoting an inclusive culture, considering the need to transform legal guidelines and pedagogical foundations into effective institutional practices. It was assumed that school inclusion requires planning, organization, and strategic leadership in the school’s daily operations. A qualitative approach of an applied nature was adopted, developed in three complementary stages: documentary analysis of Brazilian legislation on inclusive education, bibliographic review on school leadership and management, and data systematization using management tools such as SWOT analysis and the PDCA cycle. The results indicated that the school management’s performance directly influenced the consolidation of inclusive environments, especially in the organization of processes, continuous team training, and strengthening of a collaborative institutional culture respectful of diversity. Structural, pedagogical, and relational challenges that demanded planned interventions were also identified. Finally, a Practical Guide for School Managers was developed, with guidelines, protocols, and operational orientations for decision-making and the organization of inclusive processes. It was concluded that the construction of an inclusive culture depended on the management’s ability to articulate public policies, pedagogical practices, and institutional processes, consolidating inclusion as a permanent commitment in the school context.

Keywords: educational accessibility; educational equity; participatory management; strategic planning; institutional processes.

November 17, 2026

An AI artifact based on Design Thinking to support the elicitation of software requirements from user needs

The development of software oriented to user needs frequently faces challenges in requirements elicitation, stemming from ambiguities in natural language and communication failures. Given this, an intelligent artifact integrating Design Thinking and Artificial Intelligence was proposed and validated, focusing on the Empathy and Definition stages, to support the collection, interpretation, and structuring of user needs. The research, of an applied, qualitative, and exploratory nature, conducted a Proof of Concept in a controlled environment, employing hypothetical scenarios. The artifact operated through guided natural language interviews, organizing information into elements such as pain points, context, and impact, and generating traceable problem syntheses. The results indicated that the system organized unstructured reports, reduced ambiguities, and preserved the meaning of the original information in most cases. The ability to reorganize logically in complex scenarios was also observed, although with limitations in separating multiple problems and dependence on instructions. It was concluded that the artifact has potential as a support for requirements elicitation, improving the clarity and structuring of information, but requires supervision and contextual control.

Keywords: Design Thinking; Requirements Elicitation; Requirements Engineering; Artificial Intelligence; Natural Language Processing.

November 17, 2026

The Extinction of the Monophasic Tax System in the Pharmaceutical Industry in Light of the Tax Reform

The consumption Tax Reform, established by Constitutional Amendment No. 132/2023, introduced structural changes to the Brazilian tax system, with the replacement of the Social Integration Program (PIS) and the Social Security Financing Contribution (COFINS) by the Contribution on Goods and Services (CBS), in a dual Value Added Tax (VAT) model. The objective was to analyze the impacts of the extinction of the single-stage system on the pharmaceutical industry, in light of the new regulatory design and its economic and legal effects. A qualitative, exploratory research was conducted through a case study and documentary analysis of legislation and specialized doctrine. It was examined whether the end of the single-stage system would imply a relevant increase in the tax burden for the sector. The results indicated that, although the formal overcoming of the single-stage system has occurred, the material exemption of medicines has been preserved, mainly due to the application of reduced rates of 60% and, in certain cases, a zero rate of the Tax on Goods and Services (IBS) and CBS. The change in model may cause occasional variations in revenue collection, without characterizing a structural increase in the tax burden, tending towards neutrality over time. It was preliminarily concluded that the Tax Reform reorganized the taxation technique in the pharmaceutical industry, replacing a concentrated regime with more transparent exemption mechanisms. This approach proved compatible with the principles of neutrality, full non-cumulativeness, and fiscality, maintaining tax protection for essential medicines and aligning fiscal policy with the fundamental right to health.

Keywords: CBS; COFINS; IBS; Dual VAT; PIS.