Data Digitalization and Conformity Veriﬁcation in Oil and Gas Industry

Databooks Using Semantic Model Based on Ontology

Mario Ricardo Nascimento Marques Junior, Eder Mateus Nunes Gonc¸alves,

Silvia Silva da Costa Botelho and Emanuel da Silva Diaz Estrada

Center of Computational Sciences, Federal University of Rio Grande, Rio Grande, Brazil

Keywords:

Data Digitalization, Oil and Gas Industry, Ontology, Semantic Model.

Abstract:

Databooks are essential for monitoring and validating construction projects in the oil industry, containing cru-

cial information like quality certiﬁcates and technical reports. However, manual analysis of these databooks

is time-consuming, labor-intensive, and error-prone. This study proposes an intelligent system to streamline

databook search and validation, enhancing efﬁciency and accuracy. Developing a valid conceptual model for

databooks and their components presents a signiﬁcant challenge. To overcome this, we focus on acquiring

semantics for databooks and utilizing a semantic model for compliance checks. We introduce an ontology

designed speciﬁcally for verifying completeness and compliance in Brazilian oil industry documents, encom-

passing domain knowledge and veriﬁcation processes. Using the Methontology methodology, we create the

ontology and integrate it with an annotation tool to validate its ability to incorporate semantic structures and

facilitate compliance veriﬁcation. Comparative analysis with manual veriﬁcation by experts shows identical

outcomes, conﬁrming the effectiveness of the automated compliance checking process. The ontology-based

approach offers advantages such as time savings, enhanced accuracy, and simpliﬁed work for specialists. This

study contributes to oil industry document analysis by providing a semantic model that streamlines databook

veriﬁcation, with potential applications for compliance veriﬁcation of complex documents in various domains.

1 INTRODUCTION

Databooks are pivotal documents in the oil and gas

sector, housing essential equipment and work-related

data. They compile an array of information, including

engineering documents, modiﬁcations, purchase or-

ders, tests, certiﬁcates, and inspection reports (Duarte,

2010). These records encapsulate an enterprise’s his-

tory and components, potentially spanning thousands

of pages for entities like ships or entire platforms.

In the context of Brazil’s oil industry, the ABC

of Inspection of Manufacture document (S.A, 2017)

governs inspection guidelines for oil and gas equip-

ment production. As per this document, manufactur-

ing inspection is performed to verify the conformity

of equipment or materials with contractual speciﬁca-

tions at suppliers’ or sub-suppliers’ premises.

An Inspection and Testing Plan (ITP) is devised,

based on inspected equipment, technical standards,

and inspection levels. The ITP, part of the supplier’s

quality plan, outlines tests and certiﬁcations aligned

with contract-deﬁned quality and technical standards.

All equipment documents are collated in a databook,

serving as a customer’s assurance certiﬁcate and con-

taining crucial traceability information.

Post-manufacturing, a contracting party’s inspec-

tor reviews documentation for completeness and com-

pliance. Completeness entails ensuring all necessary

information, certiﬁcates, reports, and tests are present

in the databook. Compliance involves verifying that

these documents adhere to contract-speciﬁed stan-

dards. If complete and compliant, material release

occurs; otherwise, non-conforming product registra-

tion is initiated.

Professionals meticulously analyze and maintain

these databooks, checking for anomalies such as

missing reports or unsigned documents. The chal-

lenge lies in the time-intensive and costly nature of

this manual analysis. Moreover, professionals verify

databook completeness and compliance—tasks po-

tentially optimized by an intelligent search system.

Digitization and algorithms could streamline analy-

sis, aiding in data retrieval, verifying completeness,

and bolstering efﬁciency.

The majority of databook documents are scanned

as PDF images, necessitating Optical Character

180

Nascimento Marques Junior, M., Gonçalves, E., Botelho, S. and Estrada, E.

Data Digitalization and Conformity Veriﬁcation in Oil and Gas Industry Databooks Using Semantic Model Based on Ontology.

DOI: 10.5220/0012176700003543

In Proceedings of the 20th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2023) - Volume 1, pages 180-187

ISBN: 978-989-758-670-5; ISSN: 2184-2809

Document

Scan

OCR

(Optical Character Recognition)

Text file Ontology

Complete

Not complete

Figure 1: Proposal system.

Recognition (OCR) for conversion into structured text

ﬁles. However, diverse companies contribute to data-

books in varying formats, posing data modeling chal-

lenges. In this context, ontologies are more advanta-

geous than data schemas, as they provide clearer se-

mantics and mechanisms for content veriﬁcation.

Ontologies model reality and logical representa-

tions in computer science, deﬁning concepts, prop-

erties, and relationships. They enhance knowledge

sharing within speciﬁc domains through a vocabu-

lary. Works in oil and gas employ ontologies, such as

IKBO for inspection knowledge (Rachman and Rat-

nayake, 2019) and ISO 15926-based high-level mod-

els (Batres et al., 2007). DoCO (Constantin et al.,

2016) serves as a general-purpose vocabulary for aca-

demic texts.

The proposed system, as depicted in Figure 1,

transforms image ﬁles into structured Q&A formats,

leveraging an ontology for semantic inferences. This

paper’s focus is on imbuing a Q&A-structured doc-

ument with semantics for conformity veriﬁcation,

using the Methontology methodology (Fern

andez-

opez et al., 1997). The study is part of a broader

project automating veriﬁcation using data science in

a Big Data context.

The article follows this structure: Section 2 cov-

ers background and OntoToT development using

Methontology. Section 3 presents ontology incorpo-

ration of actual documents. Section 5 concludes and

discusses future work.

2 DATA DIGITALIZATION AND

CONFORMITY VERIFICATION

IN DATABOOKS

The process of digitizing data in a Databook involves

several complex steps. Firstly, the PDF ﬁle of the

Databook undergoes an OCR process. This Optical

Character Recognition (OCR) process is responsible

for converting the text embedded within images into

a machine-readable text ﬁle. By doing so, it enables

the computer to capture data from sources that were

previously inaccessible or difﬁcult to process.

However, not all the information present in a data-

book is relevant. To ﬁlter out only the pertinent in-

formation, an annotation tool has been developed.

Through this tool, experts manually select the neces-

sary information based on predeﬁned labels.

Once the relevant data has been selected, it is an-

notated in a question and answer format, following

the concepts and relationships deﬁned by the ontol-

ogy. Additionally, the annotation process generates a

ﬁle that captures spatial and geometric aspects of the

digitized document, which will be incorporated into

a dataset. This dataset will be utilized for machine

learning purposes, aiming to optimize the document

understanding and form recognition process, thereby

reducing or eliminating the need for manual annota-

tions by the user.

The labeled data can then be inserted into a

database, where it is stored and managed alongside

the OCR-digitized data. This database facilitates the

veriﬁcation of data integrity and enables retrieval and

consultation of the digitized information. The ontol-

ogy is fed with this database, allowing for inferences

related to completeness and compliance based on the

representation of concepts and relationships. The pro-

cess described above is illustrated in Figure 2.

Overall, this approach enhances the efﬁciency and

accuracy of data digitization by combining manual se-

lection, machine learning, and ontology-based infer-

ences.

2.1 OntoToT - Ontology for

Completeness and Compliance

Veriﬁcation in Petroleum Industry

Documents

Developing an ontology is an iterative process that re-

quires the use of a methodology, similar to software

development (Aminu et al., 2020). However, despite

the existence of different methodologies, there is no

widely accepted standard for ontology development

(Mendonc¸a and Almeida, 2014).

In a comparative study conducted in (Aminu et al.,

2020), the individual weaknesses of various method-

ologies are identiﬁed. Three methodologies, in partic-

ular, stand out: Methodology 101 (Noy and McGuin-

ness, 2001), Methontology (Fern

andez-L

opez et al.,

1997), and the Methodology of Gruninger and Fox

(Gr

uninger and Fox, 1995). These methodologies are

Data Digitalization and Conformity Veriﬁcation in Oil and Gas Industry Databooks Using Semantic Model Based on Ontology

181

Annotation tool

Databook OCR

Expert annotation

Text file

Dataset

Automatic form matching

Databook

Form

Database

Machine Learning

Figure 2: Databook digitalization process.

well-known and frequently used in the ﬁelds of soft-

ware engineering and knowledge representation.

The analysis reveals that some approaches pri-

marily focus on development activities, particularly

ontology implementation (as seen in Methodology

101). These approaches tend to overlook crucial as-

pects such as project management, feasibility study,

maintenance, and ontology evaluation, placing exces-

sive emphasis on implementation details (Silva et al.,

2008). It is worth noting that among these method-

ologies, Methontology is the only one that addresses

project management and maintenance phases.

For the development of the OntoToT ontology, the

Methontology methodology was chosen. Through re-

search, it was determined that this methodology is the

most suitable for the development of the ontology,

as it provides a clear deﬁnition of the development

stages and the corresponding activities. This ensures

a proper and systematic development of the ontology.

The following sections will provide a detailed expla-

nation of the main stages involved in the development

of the OntoToT ontology.

Speciﬁcation

The ontology has the representation of documents

found during the equipment inspection process as its

primary objective. In addition, the ontology must in-

clude aspects related to the completeness veriﬁcation

process of databooks.

Acquisition of Knowledge

In the knowledge acquisition phase, it was neces-

sary to read several databooks, ITPs, and purchase

orders referring to different types of equipment. In

addition, periodic meetings were held with employ-

ees of the partner company to clarify speciﬁc points

and solve possible mistrust. It is also worth noting

that its employees are specialists with several years of

experience in the ﬁeld studied here. Finally, the en-

tire knowledge acquisition process was documented

in text and slide shows.

Conceptualization

The conceptual modeling phase organizes and rep-

resents the knowledge acquired about the domain

through intermediate representations. For example, in

the ontology developed here, we used document anal-

ysis and the creation of trees to represent the concepts

and their relationships.

Formalization

The formalization activity transforms the concep-

tual model into a formal or semi-computable model

through a formal language(G

omez-P

erez et al., 2004).

When tools like the ontology editor are used, the con-

ceptualization model can be implemented directly in

several ontology languages and, consequently, for-

malization is not a mandatory step (G

omez-P

erez

et al., 2004). Thus, as the Prot

e environment will

be used to implement the ontology, this step will not

be performed.

Implementation

To develop the ontology, the Prot

e environment

(Prot

e, 2020), developed by Stanford University,

was utilized. This is an open-source tool that employs

a language based on descriptive logic (OWL-DL - On-

tology Web Language Description Logics). Prot

consists of an ontology editor and a library of plugins

with various functionalities.

For making inferences, we utilize HermiT (Her-

miT, 2020), Prot

e’s inherent inference engine tai-

lored for handling OWL-based ontologies. HermiT

adeptly checks the ontology’s coherence, identiﬁes

subsumption connections among classes, and assesses

other relevant attributes sourced from an OWL ﬁle.

Remarkably, HermiT completes these assessments

within mere seconds.

ICINCO 2023 - 20th International Conference on Informatics in Control, Automation and Robotics

182

Documentation

We also used LODE (Live OWL Documentation Envi-

ronment) (Peroni et al., 2012), an online service that

automatically generates a human-readable description

of any OWL ontology, taking into account both onto-

logical axioms and annotations and sorting them with

the appearance and functionality of a W3C recom-

mendations document. This documentation is pre-

sented to the user as an HTML page with embedded

links for easy navigation.

Ontology Description

Through the information obtained through the con-

ceptual modeling, where the knowledge about the

studied domain was organized and represented, arti-

ﬁces were generated that facilitated the ontology im-

plementation. The deﬁnition of class and subclasses

that make up the ontology is described below. The

class hierarchy is illustrated in Figure 3.

Classes Description

• PDF File: Class connected directly to Thing

class. This class represents the PDF ﬁle, the origi-

nal source of the data present in the documents of

the studied domain;

• PDF ID: Subclass of PDF File. Represents a

unique identiﬁer for each PDF ﬁle required for ﬁle

retrieval from the data storage system. Thus, ID

PDF represents an integer;

• PDF File Name: Subclass of PDF File. It is a set

of characters that aims to identify a ﬁle, and can

also be used to recover the ﬁle;

• Page: Subclass of PDF File. As a PDF can con-

tain several pages and each page can contain a dif-

ferent quality certiﬁcate, it is necessary to know

which page of the PDF this certiﬁcate is located

on. Therefore, Page represents an integer;

• Documents: Class linked directly to Thing class.

This class is responsible for grouping the types of

documents and all the attributes they can contain

in the documents of the studied domain;

• Documents Type: Subclass of Documents. It

groups the types of documents that the ontology

must include, namely: Quality certiﬁcate, Pur-

chase order and ITP.

• Quality Certiﬁcate: A quality certiﬁcate is a doc-

ument that gathers information that attests to the

quality of a given product/service according to

pre-deﬁned standards;

• Purchase Order: It is a legal contractual instru-

ment where the customer speciﬁes the equipment

to the supplier. This document has a detailed

description of the item, such as quantity, stan-

dards that must be met and type of inspection per-

formed;

• Inspection and Testing Plan - ITP: It is a doc-

ument prepared by the supplier contained in its

quality plan, following the standards established

by the quality management norms and technical

norms deﬁned in the contract;

• Document Attributes: Subclass of Documents. It

is responsible for grouping all the information that

must be extracted from the documents. Therefore,

its subclasses are Signature, Stamp, Customer,

NCM Code, Race, Date, Description, Equip-

ment, Inspection Phases, Supplier, Item, Refer-

ence Standard, Certiﬁcate Number, Purchase Or-

der Number, ITP Number, Quantity, Record, Test

Result;

• Signature: A signature or signature is a mark

or writing in some document that aims to give it

validity or identify its authorship;

• Stamp: Seal used for the recognition, proof or at-

testation of authenticity of a document;

• Client: Potential buyer or user of the supplier’s

products.

• Heat Number: It is a unique identiﬁcation code

that a technician stamps on a piece of metal to pro-

vide information about its origins;

• NCM Code: Any and all merchandise that circu-

lates in Brazil must have the NCM code (Merco-

sur Common Nomenclature) and this code must

be informed when ﬁlling in the invoice and other

foreign trade documents;

• Date: Informs the time in the calendar when a

certain document was generated;

• Description: Description of the equipment or

component being purchased through a purchase

order or evaluated through a quality certiﬁcate;

• Equipment: Equipment that is being acquired or

evaluated before an essay or test;

• Inspection Steps: Phases of the inspection pro-

cess of a certain equipment;

• Provider: Individual or legal entity that pro-

duces, assembles, creates, builds, transforms, im-

ports; exports, distributes or markets products or

services;

• Item: Code used to identify a speciﬁc piece of

equipment or component in a document;

Data Digitalization and Conformity Veriﬁcation in Oil and Gas Industry Databooks Using Semantic Model Based on Ontology

183

Figure 3: OntoToT class hierarchy.

• Reference Standard: It is a document, produced

by an ofﬁcial body accredited for this purpose,

which establishes rules, guidelines, or characteris-

tics about a material, product, process or service;

• Certiﬁcate Number: Identifying number of a

given quality certiﬁcate;

• Purchase Order Number: Identifying number of

a given purchase order;

ICINCO 2023 - 20th International Conference on Informatics in Control, Automation and Robotics

184

• ITP Number: Identiﬁer number of a given ITP;

• Quantity: Number of items in a given purchase

order;

• Register: Describes which document is the

record that a certain inspection phase was per-

formed;

• Test Result: Result of a given test or test for a

physical or chemical property;

• Type of Inspection: Indicates the type of man-

ufacturing inspection, which varies according to

the criticality of the material, operational com-

plexity of the material, complexity or novelty of

the manufacturing process and quality control,

and complexity or uniqueness of the project;

Object Properties

Object properties are basically divided into two

groups, must have and may have. During the elab-

oration of the ontology, it was argued that, in the case

of the quality certiﬁcate, some attributes must always

appear, i.e., they are mandatory for the veriﬁcation

of completeness. However, other attributes have a

certain frequency in the documents, but they are not

mandatory for verifying completeness, i.e., the qual-

ity certiﬁcate can have these attributes. In the table

1 the properties, their domains, and their ranges are

displayed.

Axioms

The axioms developed have the function of impos-

ing restrictions on the classes used for completeness

checking. That is, these axioms represent the rules

that a certain type of document must respect in order

to be considered complete or incomplete. To imple-

ment these axioms, the descriptive logic provided by

Prot

e was used. For each of the axioms a subclass

of Completeness was created. These classes are de-

scribed below.

• Class Quality Certiﬁcate Complete: A quality

certiﬁcate is said to be complete if all relevant in-

formation is correctly extracted. This information

is deﬁned by the object property Quality certiﬁ-

cate must have and are: Signature, Stamp, Cus-

tomer, Date, Certiﬁcate Number and Test Result;

• Class Quality Certiﬁcate Not Complete: A qual-

ity certiﬁcate is not complete if at least one of the

classes deﬁned in the property Quality certiﬁcate

must have is not extracted;

• Class Purchase Orders Complete: A purchase or-

der is considered complete if it has all relevant

information present. That is, a purchase order

is considered complete if the Signature, Stamp,

NCM Code, Date, Description, Supplier, Item,

Reference Norm, Purchase Order Number and

Quantity information is extracted successfully;

• Class Purchase Orders Not Complete: A pur-

chase order is considered incomplete if at least

one of the relevant information is not extracted;

• Class Complete ITPs: A ITP is said to be com-

plete if it has all relevant information present.

That is, it must contain Signature, Stamp, Cus-

tomer, Date, Equipment, Inspection Phases, Sup-

plier, ITP Number, Registration;

• Class Non-complete ITPs: A ITP is said to be

non-complete if at least one of the relevant infor-

mation is not extracted.

• ASTM A193 B7 Compliant Chemical Analysis

Certiﬁcates: For a chemical analysis certiﬁcate

to comply with the ASTM A193 B7 standard, it

is necessary that the reference standard is the one

referred to and that the values of the chemical el-

ements correspond to those shown in the Table 2;

• Non-ASTM A193 B7 Chemical Analysis Certiﬁ-

cates: For a chemical analysis certiﬁcate not to

comply with the ASTM A193 B7 standard, it is

necessary that the reference standard is the one

referred to and that at least one of the values pre-

sented in the Table 2 is different from that found

in the certiﬁcate.

2.2 Pipeline for Data Conformity

As described in this section, the data extracted

through OCR is stored in a database. Subsequently,

a Python script is utilized to query the database and

convert the data into instances within the ontology,

establishing the necessary relationships for accurate

representation.

In the Prot

e environment, the generated ﬁle,

containing only the classes and instances, is opened

along with the OntoToT ontology ﬁle, which encom-

passes all the developed classes and rules. Following

this, the inference mechanism is executed, utilizing

the axioms to perform analysis on the completeness

and conformity of the documents. The results of this

analysis are displayed on the screen. The entire pro-

cess is visually depicted in Figure 4.

3 RESULTS

To evaluate the proposed system, a set of databooks

that align with the scope of the study was chosen.

Data Digitalization and Conformity Veriﬁcation in Oil and Gas Industry Databooks Using Semantic Model Based on Ontology

185

Table 1: OntoToT object properties.

Object property Domain Range

Quality certiﬁcate

must have

Quality

certiﬁcate

Signature, Stamp, Customer, Date, Description, Certiﬁcate Number, Test Result

Quality certiﬁcate

may have

Quality

certiﬁcate

Heat number, Reference Standard, Purchase Order Number, Quantity

Purchase order

must have

Purchase

order

Signature, Stamp, NCM Code, Date, Description, Supplier, Item, Reference Standard,

Purchase Order Number, Inspection Type

ITP must have

Inspection and

Test Plan - ITP

Signature, Stamp, Customer, Date, Equipment, Inspection Phases,

Supplier, ITP Number, Registration

Database

db_to_owl.py Databook.owl

OWL

ontotot.owl

OWL

Reasoner

Conformity Not conformity

Figure 4: Ontology inference process.

Table 2: Chemical requirements requested by ASTM A193

Gr. B7.

Chemical element Range (%)

Carbon 0.37 - 0.49

Chrome 0.75 - 1.20

Sulfur max 0.040

Phosphor max 0.035

Manganese 0.65 - 1.10

Molybdenum 0.15 - 0.25

Silicon 0.15 - 0.35

As the machine learning stage is still in development,

each of these documents was manually annotated by

an expert using the annotation tool, as depicted in Fig-

ure 2. The extracted data from these annotated docu-

ments were then stored in a database.

Table 3: Veriﬁcation of compliance in chemical analysis

certiﬁcates ASTM A193 Gr. B7 manually and through the

ontology.

Certiﬁcate

number

Compliance check

by human

Compliance check

by ontology

105501 Conforming Conforming

2447/15 Conforming Conforming

1236/17 Conforming Conforming

105502 Conforming Conforming

0390/18 Nonconforming Nonconforming

Using the script illustrated in Figure 4, the data

stored in the database was used to generate instances

in the ontologies. Each generated ﬁle was subse-

quently inserted into Prot

e along with the OntoToT

ﬁle.

The inference mechanism in Prot

e was then ex-

ecuted, utilizing the axioms to provide information

about the completeness and conformity of the docu-

ments. The results of this analysis are presented in

Tables 4 and 3. For comparison, the same set of doc-

uments was analyzed by humans using the same crite-

ria, and the results obtained from this manual analysis

are also included in the tables.

4 CONCLUSION AND FUTURE

WORK

This work presents a system proposal for digitizing

and ensuring compliance in oil and gas industry data-

books. The ﬁrst step involves converting image ﬁles

into text ﬁles to enable computer understanding. To

provide a semantic structure and enable inferences,

an ontology called OntoToT was developed using

the Methontology methodology, ensuring proper de-

velopment, documentation, and incorporating project

management and maintenance stages.

The annotation process conﬁrmed that the ontol-

ogy offers sufﬁcient semantic structure for represent-

ing the analyzed databooks and demonstrated suc-

cessful integration with the annotation tool. The pro-

posed approach enables the digitization of documents,

adds semantics to the data, and facilitates automatic

ICINCO 2023 - 20th International Conference on Informatics in Control, Automation and Robotics

186

Table 4: Veriﬁcation of completeness in quality certiﬁcates performed by human and by ontology.

Certiﬁcate

Number

Completeness check

by human

Completeness check

by ontology

3418/17 Complete Complete

1049/18 Complete Complete

0700/17 Complete Complete

2650/2018 Complete Complete

111845 Complete Complete

1078/2018 Complete Complete

106210 Complete Complete

3503/2017 Complete Complete

1338739/2007 Not complete Not complete

108277 Not complete Not complete

compliance veriﬁcation, thus simplifying the work of

specialists.

Future work includes expanding the ontology to

handle more complex documentation and veriﬁca-

tions, such as assessing the completeness of an entire

databook for speciﬁc equipment. The development

of a dictionary to unify different labels for the same

attribute could also be considered. It is essential to

address ontology maintenance to accommodate new

documents and veriﬁcations effectively.

Additionally, enhancements to the annotation tool

using machine learning techniques could optimize ex-

pert annotation by requesting it only when necessary.

Improvements in the inference process could involve

developing a script for automated inference without

manual intervention, storing the results in a database.

REFERENCES

Aminu, E. F., Oyefolahan, I. O., Abdullahi, M. B., and

Salaudeen, M. T. (2020). A review on ontology

development methodologies for developing ontolog-

ical knowledge representation systems for various do-

mains. International Journal of Information Engi-

neering & Electronic Business, 12(2).

Batres, R., West, M., Leal, D., Price, D., Masaki, K., Shi-

mada, Y., Fuchino, T., and Naka, Y. (2007). An upper

ontology based on iso 15926. Computers & Chemical

Engineering, 31(5-6):519–534.

Constantin, A., Peroni, S., Pettifer, S., Shotton, D., and Vi-

tali, F. (2016). The document components ontology

(doco). Semantic web, 7(2):167–181.

Duarte, G. D. (2010). O Controle da Qualidade em Pro-

cessos de Produc¸

ao Mec

anica N

ao-Seriada. Mono-

graﬁa (Graduac¸

ao em Engenharia Mec

anica), Escola

Polit

ecnica da Universidade de S

ao Paulo, S

ao Paulo,

Brasil.

Fern

andez-L

opez, M., G

omez-P

erez, A., and Juristo, N.

(1997). Methontology: from ontological art towards

ontological engineering. In AAAI 1997. American

Asociation for Artiﬁcial Intelligence.

omez-P

erez, A., Fern

andez-L

opez, M., and Corcho, O.

(2004). Methodologies and methods for building

ontologies. Ontological Engineering: with exam-

ples from the

areas of Knowledge Management, e-

Commerce and the Semantic Web, pages 107–197.

uninger, M. and Fox, M. S. (1995). Methodology for the

design and evaluation of ontologies. In IJCAI 1995.

Citeseer.

HermiT (2020). Hermit owl reasoner. http://www.

hermit-reasoner.com/. Acess em 03/08/2020.

Mendonc¸a, F. M. and Almeida, M. B. (2014). Princ

ıpios

metodol

ogicos para desenvolvimento de ontologias:

alise das pr

aticas correntes e proposic¸

ao de melho-

rias. XV Encontro Nacional de Pesquisa em Ci

encia

da Informac¸

ao - ENANCIB.

Noy, N. F. and McGuinness, D. L. (2001). Ontology devel-

opment 101: A guide to creating your ﬁrst ontology.

Technical report, Stanford Knowledge Systems Labo-

ratory.

Peroni, S., Shotton, D., and Vitali, F. (2012). The live

owl documentation environment: a tool for the auto-

matic generation of ontology documentation. In Inter-

national Conference on Knowledge Engineering and

Knowledge Management, pages 398–412. Springer.

Prot

e (2020). Prot

e. https://protege.stanford.edu/.

Acess 20/05/2020.

Rachman, A. and Ratnayake, R. C. (2019). An ontology-

based approach for developing offshore and on-

shore process equipment inspection knowledge base.

In International Conference on Offshore Mechan-

ics and Arctic Engineering, volume 58783, page

V003T02A084. American Society of Mechanical En-

gineers.

S.A, P. B. (2017). ABC da inspec¸

ao de fabricac¸

ao. Instru-

mento informativo, Petr

oleo Brasileiro S.A, Brasil.

Silva, D. L. d., Souza, R. R., and Almeida, M. B. (2008).

Ontologias e vocabul

arios controlados: comparac¸

de metodologias para construc¸

ao. Ci

encia da

informac¸

ao, 37:60–75.

Data Digitalization and Conformity Veriﬁcation in Oil and Gas Industry Databooks Using Semantic Model Based on Ontology

187