Article

October 13, 2026

Construction of a geological data catalog to enhance data science and analytics

Construction of Geological Data Catalog to Empower Data Science and Analytics

Márcio Brito Moraes; Anna Carolina Martins

DOI: 10.22167/2675-6528-202603257

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

The dispersion, structural and semantic heterogeneity, and the lack of standardization of geological data hinder its discovery, understanding, integration, and reuse in analytical workflows in Oil and Gas Exploration Geology. With the aim of strengthening information governance and enhancing the use of Data Science and Analytics in this context, a Geological Data Catalog was built. The methodology employed combined literature review, analysis of standards and best practices, use of open data from the National Agency of Petroleum, Natural Gas and Biofuels (ANP), conceptual modeling, metadata specification, and implementation of a navigable prototype in Power BI. As results, a conceptual model of the catalog, a metadata dictionary with canonical attributes, controlled vocabularies for critical fields, a traceability layer between original sources and catalog elements, and a quality layer with rules, summarized findings, and detailed violations were consolidated. The prototype demonstrated that a structured catalog can reduce ambiguities, increase transparency about provenance and quality, and create more favorable conditions for the preparation and analytical use of geological data. It was concluded that the adoption of a geological data catalog, supported by standardized metadata and quality mechanisms, constitutes a viable and useful approach to strengthen information governance and enhance data science and analytics applications in geoscientific contexts.

Keywords: Digital curation; Information governance; Interoperability; Metadata; Information quality.

1. Introduction

The advancement of Data Science and Analytics technologies has driven digital transformation in data-intensive organizations, notably in the oil and gas industry. Specifically in the technical areas of Exploration Geology, the strategic use of high-quality geological data enables greater efficiency in generating geoscientific knowledge. This enhances decision-making processes and reduces uncertainties associated with basin assessment, reservoir modeling, and the identification of new exploratory opportunities.

However, the dispersion, structural and semantic heterogeneity, and the absence of standardization of geological data still represent significant barriers to the application of advanced analysis and artificial intelligence techniques. The data are often distributed across multiple systems and spreadsheets, presenting inconsistent nomenclatures and a lack of metadata describing their origin, context, and quality.

In this scenario, the construction of a Geological Data Catalog emerges as a strategic initiative to organize, document, and standardize the metadata associated with various geological disciplines, such as stratigraphy, geochemistry, sedimentology, structural geology, and rock and fluid analyses. This type of catalog acts as a curation and governance tool, facilitating the discovery, understanding, and reuse of data by different technical teams and analytical systems.

The concept of a data catalog was adopted as the central element of this research. According to Barbieri (2020), a data catalog constitutes a structured repository that organizes, documents, and describes informational assets through standardized metadata, enabling their identification, understanding, traceability, and reuse. Complementarily, Batini and Scannapieco (2016) highlight that data catalogs are essential instruments for ensuring information quality, consistency, and reliability, especially in environments characterized by a large volume and diversity of data. In the context of data science and analytics, catalogs play a strategic role by enabling the efficient discovery of datasets, reducing information asymmetries between producers and consumers, and increasing transparency regarding the origin, quality, and limitations of available data ([DAMA] International, 2021).

Applied to the geological domain, this role intensifies due to the multidisciplinary nature of the data, spatial and temporal complexity, and the multiplicity of formats and scales. A structured catalog tends to strengthen governance, promote interoperability between domains, and create favorable conditions for the intensive use of analytics in exploratory activities. Thus, the research is justified by the need to overcome the barriers imposed by the disorganization and lack of standardization of geological data, seeking to enhance the value generated by data science initiatives. The objective of this study is to propose a conceptual model of a Geological Data Catalog for curation, standardization, description, and integration of metadata in Oil and Gas Exploration Geology, aiming to strengthen governance, information management, and the use of Data Science and Analytics.

2. Material and Methods

The research was characterized as a study of an applied nature, with a qualitative and exploratory approach, aimed at building a Catalog of Geological Data. This approach was adopted with the objective of producing practical and applicable knowledge for the organization, standardization, and analytical use of technical data in Oil and Gas Exploration Geology, seeking to address data quality dimensions such as completeness, consistency, accuracy, validity, and integrity (Gil, 2019).

The methodological strategy articulated a literature review, metadata analysis, conceptual modeling, standardization of documentary structures, prototype implementation, and functional evaluation. The study was structured in sequential and incremental phases, covering the literature review and scope definition, conceptual modeling and metadata definition, detailed catalog specification and traceability mapping, and prototype implementation with functional validation.

In the first phase, a systematic literature review was conducted, which included technical books, scientific articles, international standards, and institutional documents. Recognized references on data governance, information quality, data catalogs, metadata, and geological data management were prioritized, including Barbieri (2020), Batini and Scannapieco (2016), and the [DAMA-DMBOK] ([DAMA] International, 2021).

This review allowed the consolidation of the conceptual and normative foundations that guided the metadata analysis and the development of the catalog’s conceptual model. The standards ([ISO] 19115:2014) and ([ISO] 19157:2013), as well as the guidelines ([INSPIRE] Metadata Implementing Rules, 2007), were adopted to structure metadata and address the quality of geoscientific and geospatial data.

The delimitation of the scope and sources of the conceptual geological catalog was conducted from the analysis of public open data bases from the National Agency of Petroleum, Natural Gas and Biofuels (ANP, 2020). Metadata from domains relevant to Oil and Gas Exploration Geology, accessible through the official ANP portal, were consumed.

The selected domains included Wells, Core, Testimonials, Sidewall Samples, Plugs, Slices, Cells, Fluids, Geochemical Surveys, and Geochemical Data. The choice of these domains was motivated by their representation of recurring technical objects in geological disciplines and by their ability to structure a metadata-based catalog, aligned with data management and governance principles.

In the next phase, the conceptual modeling of the geological data catalog was developed, including the identification of entities, attributes, relationships, and data domains. This modeling was documented through an Entity-Relationship Diagram ([ERD]) and a data dictionary, following good conceptual modeling practices recommended by [DAMA-DMBOK] ([DAMA] International, 2021).

The [DER] was built using dbdiagram.io, an online tool from Holistics Software (HOLISTICS, 2026), which allows defining and documenting database schemas. The [DER] represented the conceptual structure of the catalog, highlighting central entities, supporting entities, and provenance and contextualization links between wells, samples, and geochemical data.

A Data Dictionary was created to describe, in a standardized and systematic way, the data elements used, detailing meaning, type, examples, and rules associated with each attribute. This artifact acted as an operational reference for understanding the content of each domain, guiding the standardization of metadata, and facilitating data discovery and reuse.

The conceptual framework was organized from four central domains: Well, Samples (encompassing Chute, Cores, Sidewall Cores, Plugs, Thin Sections, Cells, and Fluids), Geochemical Survey, and Geochemical Data. The selection of essential metadata was conducted systematically to enable discovery, understanding, traceability, relationship, and interoperability between geological domains.

The main entities were defined as: Well, Sample, Geochemical Survey, and Geochemical Data. Supporting entities, such as Sedimentary Basin, Stratigraphic Unit, Project/Program, Location, Laboratory/Executing Entity, and Document, were included to ensure traceability, contextualization, and organization of data assets.

Conceptual relationships were established to reflect the minimum integration chain. A Well was associated with a Sedimentary Basin and could contain Core Samples, Cores, and Sidewall Samples. Cores could originate Plugs, and Thin Sections and Mounts could be associated as derived from Core, Cores, Sidewall Samples, and Plugs.

In parallel with conceptual modeling, the identification and standardization of essential metadata to characterize geological data were conducted, taking the analyzed ANP databases as a direct basis. The mapping of sources and their original schemas, the inventory of technical, spatial, temporal, and administrative attributes, and semantic classification were carried out.

A central methodological component was the traceability mapping between the original database fields and the canonical elements of the catalog. This mapping was designed to clarify the correspondence between original sources and catalog documentary structures, associating original names with standardized names, interpretation rules, and links to value domains.

Additionally, semantic enrichment was performed by defining controlled vocabularies for critical categorical attributes, especially in cases where codes, acronyms, or technical classifications could generate ambiguities of interpretation. This procedure strengthened the catalog’s consistency and made its elements more suitable for reuse in multidisciplinary and analytical contexts.

The prototype implementation was carried out in Power BI. The choice of the tool was due to its ability to combine, in the same environment, tabular modeling, definition of relationships, creation of analytical measures, and construction of interactive interfaces. This allowed the catalog to be materialized in an interactive analytical environment without the need to develop a complete web application.

The prototype was structured from metadata tables, including datasets, columns, vocabularies, allowed values, relationships, and quality rules. Pages were implemented for catalog overview, dataset exploration, field exploration, attribute detailing, vocabulary visualization, conceptual relationship representation, and data quality monitoring.

Functional validation focused on verifying the consistency between the defined conceptual structures and their behavior in the implemented prototype. The consistency of relationships between tables, the functioning of navigation between pages, the behavior of filters and segmentations, the association between fields and vocabularies, the display of allowed values, and the consistency of the quality layer were observed.

The study did not involve direct data collection from people, nor the conducting of interviews, questionnaires, or any form of interaction with human participants. All data used were obtained exclusively from ANP’s open data, available for access on the internet. Therefore, it was not necessary to submit the project to the Research Ethics Committee (CEP).

3. Results and Discussion

The analysis performed confirmed that the construction of a Geological Data Catalog, based on standardized metadata, structured conceptual modeling, and a navigable prototype, represents a viable and useful approach to strengthen the governance, curation, and use of Data Science and Analytics in Oil and Gas Exploration Geology. The investigation showed that, although the analyzed datasets presented relatively standardized documentation at the dataset level and attribute level, structural and semantic heterogeneities persisted that limited the discovery, understanding, integration, and analytical reuse of the data. This finding is in line with the data governance and management literature, which emphasizes the need to organize informational assets into structures that clarify meaning, context, provenance, and quality (Barbieri, 2020; [DAMA] International, 2021; Batini and Scannapieco, 2016).

The characterization of the databases revealed a level of standardization in the dataset and attribute metadata, with all analyzed domains documented under the structures of “Additional Information” and “Details about each column of the dataset”. This initial pattern is a positive starting point, indicating the existence of a minimum documentary core to support cataloging initiatives. However, significant heterogeneity was observed in the number of attributes and naming patterns, especially in the Wells domain, which presented more columns and distinct technical conventions compared to the samples bases, which frequently used names with accents, spaces, and line breaks. This contrast highlighted the need for syntactic and semantic normalization to improve interoperability, intelligibility, and reuse, aligning with the role of data catalogs in information management (Barbieri, 2020; [DAMA] International, 2021; Batini and Scannapieco, 2016).

From a semantic point of view, the research identified important challenges for standardization, including documentation gaps and enrichment opportunities. In several sample domains, a single global identifier per record was not found, which led to the need to use composite keys and suggested identifiers to ensure uniqueness and provenance in the catalog. Furthermore, redundant attributes were observed across multiple domains, such as sedimentary basin, reservoir, and geographic context information, indicating the need for harmonization. Crucial fields for analysis, such as depths, volumes, and intervals, were often described as text in the original metadata, requiring the explicit definition of normalized types, units of measure, and validation rules. These findings corroborate the dimensions of completeness, consistency, validity, and integrity highlighted by Batini and Scannapieco (2016) and by [DAMA-DMBOK], as well as the emphasis of [ISO] 19157:2013 on the formalization of quality criteria for geospatial and geoscientific data ([DAMA] International, 2021; [ISO] 19157:2013 Geographic Information Data Quality, 2013).

Regarding the relationships between samples and their derivatives, the analysis confirmed that the metadata partially supported the hierarchy defined in the study. Samples such as Chute, Cores, Sidewall Samples, and Fluids were associated with Wells; Plugs were identified as derivatives of Cores; and Slices and Cells as derivatives of Chute, Cores, Sidewall Samples, and Plugs. However, it was found that this derivation was not always explicitly stated by formal origin identifiers, which imposed restrictions on complete traceability. The link between Plugs and Cores was represented by the core number, but the derivation relationships of Slices and Cells depended on a categorical attribute of the source sample type, without an unequivocal reference to a specific parent sample. In response to this limitation, the catalog modeling incorporated enrichment mechanisms, such as composite keys, suggested identifiers, and explicit provenance attributes, including reference to the source sample, derivation type, and standardized interval. This result is consistent with the literature, which points out that the absence of provenance metadata and formal consistency rules increases the risk of divergent interpretations and reduces the potential for reliable data reuse (Batini and Scannapieco, 2016; [ISO] 19157:2013 Geographic Information Data Quality, 2013).

Despite these limitations, the consolidated conceptual model proved sufficient to serve as the basis for the prototype implementation. The main contribution of this stage was to demonstrate that the analyzed domains could be organized into a coherent structure of entities, attributes, and relationships, capable of representing, at a catalog level, the main informational objects of Exploration Geology. The Entity-Relationship Diagram (ERD) produced synthesized this structure, clarifying both the central entities and the fundamental links of integration and provenance between wells, samples, and geochemical data. This ERD, built using the dbdiagram.io tool, graphically represented the entities Well, Samples (Chip, Cores, Sidewall Cores, Plugs, Slices, Cells, and Fluids), Geochemical Survey, and Geochemical Data, as well as support entities such as Sedimentary Basin and Stratigraphic Unit, detailing their attributes and relationships with cardinalities, which is fundamental for understanding the scope and standardization of concepts.

Based on the conceptual model, the detailed specification of the catalog and metadata dictionary was consolidated. This structure began to explicitly state, for each catalog attribute, the original column name, the canonical name, the source data type, the normalized type, the unit of measure (when applicable), the mandatory nature, the role in the relationship, the related key, the value domain, and the validation rules. This specification significantly expanded the catalog’s utility for Data Science and Analytics, offering users not just a list of fields, but an interpretable set of useful information for data selection, preparation, and validation. In analytical workflows, this approach reduces semantic ambiguity, facilitates understanding of attribute meaning, and aids in anticipating problems with typing, units, or compliance. The Data Dictionary synthesized from the data sources consolidated the most relevant attributes for cataloging, relationship, and interoperability, acting as an operational reference for understanding the content of each domain.

At the dataset level, the research confirmed that all analyzed domains presented a common set of eight metadata fields: title, description, data origin, data source, file format, update frequency, notes, and contact. This initial pattern favored the cataloging stage, providing a minimal identification base. However, critical analysis showed that, for Data Science and Analytics purposes, this structure was insufficient, as it did not consistently specify aspects such as temporal and spatial coverage, dataset version, extraction date, terms of use, lineage, and quality indicators. In response, the study proposed expanding the dataset metadata template, incorporating additional fields focused on discovery, fitness-for-use assessment, and interoperability, in alignment with the recommendations of [DAMA-DMBOK], [ISO] 19115:2014, and the [INSPIRE] guidelines ([DAMA] International, 2021; [ISO] 19115:2014 Geographic Information Metadata, 2014; [INSPIRE] Metadata Implementing Rules, 2007). This expansion is particularly relevant in analytical environments, where the choice of a dataset depends not only on its nominal content but also on its temporal and spatial context and its documentary reliability.

At the column level, the results reinforced the need for a canonical documentation layer. Although all domains presented the four original attribute documentation fields (column name, example, data type, and description), the research confirmed significant weaknesses, such as heterogeneous field names, metric fields described as text, lack of unit standardization, and descriptive gaps. The final catalog specification addressed these issues by incorporating additional documentary attributes, such as normalized type, unit of measure, mandatory status, validation rule, and relational role. In practice, this layer has direct implications for Analytics: attributes such as depth, volume, top, and base become more consistently interpretable and comparable; categorical fields become associated with controlled domains; and the preparation of datasets for analytical modeling requires less manual interpretation effort. This result reinforces the idea that a data catalog, when well-specified, acts as an infrastructure for understanding and not just as a passive inventory of schemas.

The research also confirmed that semantic standardization requires controlled vocabularies for critical categorical attributes. Among the most relevant cases, plug orientation, source sample type, and classifications related to geochemical fluids and data stood out. This need was operationalized in the prototype through specific vocabulary structures and allowed values, linked to the corresponding catalog fields. This result adds concrete value because it transforms previously isolated codes and categories into elements belonging to a formal domain of valid meanings. In a geological context, this reduces ambiguities, facilitates interdisciplinary integration, and supports consistent data reuse, especially when different teams or analytical users do not share the same detailed technical repertoire. The prototype’s controlled vocabularies page allowed visualization of the defined value domains, their valid values, and the association of these vocabularies with the documented fields.

The implementation of the prototype in Power BI materialized the conceptual model and the documentary specification in a navigable environment, demonstrating the applicability of the proposal. The choice of the tool was due to its ability to combine, in the same environment, tabular modeling, definition of relationships, creation of analytical measures, and construction of interactive interfaces. This allowed for the rapid construction of a demonstrative artifact with low development complexity and high capacity for communicating the result. The prototype’s homepage was designed to offer the user a synthetic view of the catalog and organize access to the different exploration layers implemented, such as datasets, fields, vocabularies, relationships, and quality. This interface translated the work’s proposal into a concrete user experience, transforming the catalog into a portal for discovering and understanding informational assets.

The ability to support the discovery of available datasets is one of the most basic and important functions of a catalog. In the context of this research, this need was particularly relevant, as the analyzed domains have distinct purposes, different granularities, and varied technical applications. The creation of a specific page for dataset exploration allowed for navigation structuring at the dataset level, offering the user a mechanism to quickly locate the domains most aligned with their analytical interest and to understand, in a summarized way, their documentary context. This functionality is essential to reduce the effort of locating and screening sources even before a more detailed attribute analysis. Once a relevant dataset is identified, the next step in using the catalog consisted of understanding the documentary context of that dataset in greater depth, for which a detail page was structured that gathers essential information about description, origin, periodicity, format, contact, and other dataset-level metadata, in addition to connecting the dataset to the documented attribute layer. This functionality proved particularly important because the analytical use of a dataset depends not only on identifying its existence but also on its contextual interpretation, including origin, scope, and update conditions. Thus, the dataset detailing reinforced the catalog’s role as an instrument for understanding and not just an inventory.

If dataset-level exploration is important for source discovery, field-level exploration is decisive for analytical use itself. In Data Science and Analytics practices, the search for relevant variables is often a critical step, as it requires quickly identifying which attributes contain the necessary information, in which domains they appear, and under what semantic and technical conventions they are documented. The field exploration page was structured to meet precisely this need, offering a navigable view of the catalog’s attribute layer. With this, the prototype began to support not only the discovery of datasets but also the guided selection of variables based on documentary and semantic criteria. The identification of a relevant field does not end the need for documentation; on the contrary, it opens space for an even more important step: a deep understanding of the selected attribute. In the prototype, this function was implemented through a specific detail page, which concentrates elements such as original name, canonical name, data type, unit, relational role, validation rule, example, and associated vocabulary. This structure proved particularly valuable for supporting data preparation, cross-domain integration, and analytical interpretation, as it reduced semantic ambiguities and made explicit information that, in many scenarios, would remain implicit or scattered across different documentary sources. Thus, the field detail page concretely expressed the catalog’s role as a layer of technical and semantic understanding.

One of the central contributions of the proposed catalog consisted of explicitly stating, in a structured way, the links between entities and domains that, in the original databases, did not always appear clearly or formally enough. The prototype’s relationships page was developed to make this structure more understandable to the user, bringing conceptual modeling closer to the navigation experience. By presenting the links between wells, samples, geochemical data, and other domain elements, this interface reinforces the importance of contextualization and provenance for reliable data use. In other words, it shows that the catalog is not limited to the isolated description of tables and fields, but also incorporates the relational logic that supports integration and traceability. The prototype’s relationships page translated the catalog’s conceptual structure into a visualization to support the interpretation of links between entities and domains.

Semantic standardization was one of the most important aspects of catalog construction, especially in handling categorical attributes whose meaning depends on codes, acronyms, or technical classifications. To operationalize this dimension, the prototype incorporated a specific controlled vocabularies page, allowing visualization of defined value domains, their valid values, and the association of these vocabularies to the documented fields. This functionality added value because it reduced interpretation ambiguities, strengthened semantic consistency, and facilitated data reuse by users with different levels of familiarity with the geological domain. In the context of the prototype, the vocabularies page represented the practical materialization of the proposed formal treatment of value domains, displaying the allowed values and their associations to the documented fields.

In addition to documentation and metadata navigation, the prototype incorporated a layer focused on visualizing the quality of the cataloged data. This expansion proved relevant because, in Data Science and Analytics contexts, data discovery needs to be accompanied by some degree of transparency about its fitness for use. The quality overview page was designed to gather synthetic indicators, occurrence distribution, and a first layer of detail on affected records, allowing the user to quickly perceive where the main documented problems are concentrated. This functionality reinforced the idea that the catalog can operate as a governance instrument, not just an inventory. In addition to an aggregated quality view, the research showed the importance of analyzing documented problems according to the specific logic of each rule. This perspective allows for a more precise understanding of which types of occurrences are concentrated in certain datasets, which quality dimensions are most impacted, and how the criticality of the findings is distributed across the catalog. The quality-by-rule page was structured to offer this more analytical reading, functioning as a complement to the overview and allowing for a deeper interpretation of the cataloged problems. With this, the prototype began to bring together, in a single solution, both an executive view of quality and a view oriented towards detailed analysis, where occurrences can be explored by rule, criticality, quality dimension, and dataset.

The critical discussion of the results confirms that the catalog proposal addresses the research problem formulated in the Introduction. The prototype implementation demonstrated that the construction of a structured catalog is capable of reducing barriers to discovery, understanding, and reuse of geological data, especially when combined with metadata standardization, provenance explicitation, the use of controlled vocabularies, and the incorporation of quality mechanisms. In terms of Data Science and Analytics, this contribution is particularly relevant because working with geological data depends not only on the existence of files, but on understanding their technical context, their usage restrictions, and their documentary reliability. A data scientist or analyst wishing to select variables to compare samples by depth, cross-reference geochemical information by well, or evaluate relationships between sample type and laboratory attributes directly benefits from a catalog that already explicits expected types, units, vocabularies, and minimum integrity rules.

At the same time, the results also evidenced important limitations. The main one continues to be the dependence on the quality and completeness of the metadata originally provided by the analyzed source. In domains where there are no universal identifiers or explicit formal relationships between samples and derivatives, the catalog relies on additional enrichments to represent traceability more robustly. Nevertheless, these limitations do not weaken the proposal; on the contrary, they reinforce its relevance, as they make visible gaps that, without a structured catalog, would remain implicit and hinder the reliable use of data. In summary, the results and discussions of this research demonstrate that the Geological Data Catalog was not only modeled but also specified, implemented, and validated at the prototype level. The study confirmed the importance of metadata standardization, provenance explicitation, the use of controlled vocabularies, and the incorporation of quality mechanisms as central components of a catalog oriented towards governance and analytical use. The developed prototype does not aim to exhaust the complexity of the geological domain nor to replace a future corporate solution, but it concretely demonstrated that it is possible to structure a useful artifact for discovery, understanding, curation, and support for Data Science and Analytics initiatives in geoscientific contexts.

4. Conclusion

The study aimed to propose a conceptual model of a Geological Data Catalog for curation, standardization, description, and integration of metadata in Oil and Gas Exploration Geology, strengthening governance, information management, and the use of Data Science and Analytics. It was found that the dispersion and structural and semantic heterogeneity of geological data represented significant barriers, which were addressed by constructing a robust conceptual model. The need for syntactic and semantic normalization, as well as the importance of controlled vocabularies and traceability mechanisms to overcome documentation gaps and inconsistencies, was identified. The implementation of a navigable prototype in Power BI demonstrated the feasibility of the proposal, consolidating a metadata dictionary with canonical attributes and a quality layer. This approach contributed to reducing ambiguities, increasing transparency about data provenance and quality, and creating more favorable conditions for the preparation and analytical use of geological information, enhancing the value generated by data science initiatives.

Despite the advances, the dependence on the quality and completeness of the metadata originally provided by the analyzed sources was observed as a limitation. In domains without universal identifiers or explicit formal relationships between samples and derivatives, the catalog required additional enrichments to robustly represent traceability. However, these restrictions do not invalidate the proposal; on the contrary, they highlight the relevance of the catalog as a tool to make visible and manageable gaps that would otherwise remain implicit and hinder the reliable use of data. It is concluded that the adoption of a geological data catalog, based on standardized metadata and quality mechanisms, constitutes a useful and viable approach to strengthen information governance and drive data science and analytics applications in geoscientific contexts. The prototype, although it does not replace a complete corporate solution, has established a concrete basis for the discovery, understanding, and curation of informational assets.

Bibliographic References

[ANP], Agência Nacional do Petróleo, Gás Natural e Biocombustível. Dados abertos. 2020. Disponível em: <https://www.gov.br/anp/pt-br/centrais-de-conteudo/dados-abertos>. Acesso em 04 de fevereiro de 2026.

[DAMA] International. 2021. [DAMA-DMBOK]: Data Management Body of Knowledge. 2ed. Technics Publications, Bradley Beach, NJ, EUA.

Barbieri, C. 2020. Governança de Dados. Alta Books, Rio de Janeiro, RJ, Brasil.

Batini, C.; Scannapieco, M. 2016. Data and Information Quality: Dimensions, Principles and Techniques. Springer, Cham, Suíça.

European Commission. 2007. [INSPIRE] Metadata Implementing Rules: Technical Guidelines based on EN [ISO] 19115 and EN [ISO] 19119. Version 1.2. European Commission, Brussels, Bélgica.

Gil, A.C. 2019. Métodos e técnicas de pesquisa social. 7ed. Atlas, São Paulo, SP, Brasil.

Holistics Software Pte Ltd [HOLISTICS]. 2026. dbdiagram.io: ferramenta online para construção de [DER]. Disponível em: <https://dbdiagram.io/>. Acesso em: 04 de fevereiro de 2026.

International Organization for Standardization [ISO]. 2013. [ISO] 19157:2013 Geographic information — Data quality. [ISO], Geneva, Switzerland.

International Organization for Standardization [ISO]. 2014. [ISO] 19115:2014 Geographic information – Metadata. [ISO], Geneva, Switzerland.

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

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.