Local factors in accident analysis in the Brazilian offshore oil and gas industry

Abstract

Objective  This study applies Muschara’s local factors classification to the analysis of three accidents in Brazil’s offshore oil and gas industry: FPSO Cidade de São Mateus (2015), Platform P-53 (2019), and Platform P-19 (2022).

Methods  Using a qualitative and descriptive approach, the research relied on official investigation reports from the Brazilian National Agency of Petroleum, Natural Gas and Biofuels (Agência Nacional do Petróleo, Gás Natural e Biocombustíveis, ANP).

Results  Total of 38 root causes were identified and categorized into external and internal local factors influencing human performance. External factors predominated, mainly related to communication, feedback, resources, and organizational incentives, exposing systemic weaknesses that contributed to the events.

Conclusion  Considering the identified factors, including the organizational factors that influence human performance and support operational reliability in offshore environments, it is reinforced the need to strengthen the systemic approach to accident investigation within Brazil’s Operational Safety Management System (Sistema de Gerenciamento da Segurança Operacional, SGSO).

Keywords
Human Factors Engineering; Occupational Accidents; Oil and Gas Industry; Industrial Safety; Occupational Health

Resumo

Objetivo  Este estudo aplica a classificação de fatores locais, proposta por Muschara, à análise de três acidentes na indústria de petróleo e gás offshore brasileira: FPSO Cidade de São Mateus (2015), Plataforma Petrobras 53 (2019) e Plataforma Petrobras 19 (2022).

Métodos  A pesquisa, de caráter qualitativo e descritivo, baseou-se na análise documental de relatórios oficiais da Agência Nacional do Petróleo, Gás Natural e Biocombustíveis (ANP).

Resultados  Foram identificadas 38 causas-raiz, classificadas entre fatores locais externos e internos, que influenciaram o desempenho humano. Houve predominância de fatores externos relacionados à comunicação, ao feedback, aos recursos e incentivos organizacionais, evidenciando fragilidades sistêmicas que contribuíram para os eventos.

Conclusão  Considerando os fatores identificados, inclusive organizacionais que moldam o desempenho humano, e a confiabilidade operacional no setor offshore brasileiro, destaca-se a importância de fortalecer a abordagem sistêmica da investigação de acidentes no âmbito do Sistema de Gerenciamento da Segurança Operacional (SGSO).

Palavras-chave
Engenharia Humana; Acidentes do Trabalho; Indústria de Petróleo e Gás; Segurança Industrial; Saúde do Trabalhador

Introduction

Socio-technical systems can be defined as complex configurations made up of the continuous interaction between human, technological and organizational components, whose performance depends on their ability to adapt to the variations and uncertainties of the operational context¹. The Brazilian offshore oil and gas industry is an example of this type of system, as it integrates highly critical equipment, dynamic processes and operational procedures that require coordination between management structures. Understanding human action is essential not only to avoid accidents, but also to promote reliability and adaptability in organizations operating in high-risk contexts.

The National Agency of Petroleum, Natural Gas and Biofuels (Agência Nacional do Petróleo, Gás Natural e Biocombustíveis, ANP) is the body responsible for overseeing oil and natural gas exploration and production activities, in accordance with the provisions of Law No. 9,478/1997², and aims to prevent failures in the operational safety of facilities and minimize possible damage to life, the environment and property.

In September 2023, Brazil produced around 3.672 million barrels of oil per day and 157,989 thousand cubic meters of natural gas per day, totaling 4.666 million barrels of oil equivalent per day³, reflecting the importance of the Brazilian offshore sector for the country’s economy. Maritime operations were responsible for around 97.7% of oil production and 85.8% of gas production in Brazil, accounting for 67.1% of the hours worked in the oil sector⁴.

Through Resolution 43/2007, the ANP instituted the Operational Safety Management System (Sistema de Gerenciamento da Segurança Operacional, SGSO), establishing requirements for the implementation of formal safety management systems on offshore drilling and production units⁵. This regulatory framework brought Brazil closer to international risk management practices and consolidated the responsibility of operating companies for the integrity of their systems and processes. Later, ANP Resolutions 851/2021 and 882/2022 increased the emphasis on organizational culture, accident investigation, and continuous improvement of safety management processes⁶,⁷.

Despite the regulatory advances, the predominant focus of investigations in the country is still based on a traditional conception of safety, aligned with the concept of Safety-I⁸. From this perspective, safety is defined as the absence of failures, and accidents are explained as linear chains of causes and effects, often attributed to human error. This view is limited in complex systems, in which interactions between people, processes and technologies generate behaviors that don’t fit into deterministic models⁸,⁹,¹⁰.

Since 2010, a paradigm shift in the understanding of safety in complex systems has been consolidated, driven by authors such as Erik Hollnagel, Sidney Dekker, and René Amalberti⁸,⁹,¹⁰. Defined as Safety-II, this approach proposes the definition of safety as the ability of a system to succeed under variable conditions, emphasizing the role of human and organizational adaptations in its performance. This approach recognizes that the same set of practices and decisions that lead to successful results can also, under certain circumstances, contribute to failures, shifting the focus from analyzing what went wrong to understanding how the work is actually done and not just how it is prescribed.

In dialogue with Safety-II, Resilience Engineering proposes a theoretical framework that seeks to understand how highly complex and reliable organizations maintain control of their operations in the face of uncertainty. According to Hollnagel, resilience is defined by a system’s ability to anticipate, monitor, respond to and learn from errors, and above all to learn from everyday situations in which no error or deviation has been identified¹¹. This approach relates human error to performance variability and the need for systems capable of absorbing and adapting to operating conditions¹².

The term “human factors” was developed with a focus on adapting tasks, equipment, and environments to the physiological and cognitive characteristics ofoperators¹³. Over time, this concept has evolved into a systemic approach, which considers the interactions between people, technology, processes, and organizational structures as determinants for the safe and effective performance of socio-technical systems¹⁴.

In this sense, human factors are not restricted to individual attributes but encompass collective and contextual dimensions that directly influence workers’ decisions and actions¹⁴,¹⁵. Therefore, the evaluation of human error is no longer treated as an isolated cause in incidents, but as an indication of broader systemic dysfunctions¹⁵.

In Brazil, the human factor as a structuring element of operational safety was consolidated in 2023, with the publication of Technical Note No. 10/2023/SSO/ANP¹⁶. This document, with technical references from scientific literature, guides the integration of human factors into the SGSO and recommends the use of reports from the International Association of Oil and Gas Producers (IOGP), especially Report 454, which focuses on human factors engineering in projects, and Report 621, which focuses on the investigation of events and the role of human and organizational behavior¹⁷,¹⁸. Report 621 reinforces that when an event is attributed to human error, the investigation must continue until it identifies the systemic conditions that contributed to that behavior, avoiding causal simplification and punitive bias.

It is in this context that the classification of accidents proposed by Muschara¹⁹, developed from the principles of the Human Performance Improvement (HPI) culture adopted in the nuclear industry, fits in.

HPI is a systemic approach aimed at the continuous improvement of organizational results, based on the premise that human performance is influenced by the context in which work is carried out. As such, it focuses on identifying and addressing the root causes of performance problems, especially those related to organizational, environmental and managerial factors, to the detriment of explanations centered exclusively on individual failings. This perspective reinforces the understanding of accidents as multifactorial phenomena, embedded in complex socio-technical systems²⁰.

Muschara’s¹⁹ classification is based on the analysis of so-called local factors, which he defines as conditions present in the work environment that directly influence human performance. These factors represent elements of the system that shape the way people perceive, decide, and carry out their activities. The author¹⁹ organizes them into external local factors and internal local factors, distinguishing, respectively, those belonging to the organizational environment and those linked to the individual characteristics of workers.

External local factors correspond to the structural, organizational and contextual conditions that affect work. They include: (i) Requirements, Expectations, and Feedback, related to the clarity of goals, operational guidelines, and the feedback received on performance; (ii) Tools and Resources, which cover the availability and suitability of instruments, equipment, systems, and infrastructure necessary for the safe execution of tasks; (iii) Incentives and Disincentives, referring to formal or informal mechanisms that direct behavior, such as production pressures, rewards, punishments, or the prioritization of operational goals over safety requirements¹⁹.

Internal local factors involve individual capabilities, personal attributes, and elements intrinsic to the worker. Muschara¹⁹ describes three main groups: (iv) Knowledge and Skills, which comprise the training, experience, technical competence and practical mastery needed to deal with routine and anomalous operating conditions; (v) Capacity, which includes physical, cognitive, and emotional limitations and variable human conditions such as fatigue, overload, stress, and functional limitations that can affect the safe execution of a task; and (vi) Expectations, Motivation, and Personal Preferences, which influence the way each individual perceives risk, prioritizes tasks and responds to system demands.

Although the classification adopted separates internal and external factors for analytical purposes, it is recognized that, as proposed in the literature on human factors and ergonomics¹⁵, internal factors often reflect or result from the organizational conditions of the system (external factors) and should therefore be interpreted in their interdependence with the work context.

Local factors play a critical role in the behavior of workers and the effectiveness of operations in complex production systems such as offshore oil installations. These factors, both internal and external, interact directly with the culture and work environment, affecting decision-making, individual performance, and collective safety¹⁹.

Reason, in 1997²¹, already pointed out that accidents in the workplace, for the most part, can be induced in workers by the system, for example, by confusing procedures, lack of signage, and unmaintained sensors.

With the above in mind, the general objective of this study was to apply the classification proposed by Muschara in 2018¹⁹, with a focus on identifying and interpreting the local factors associated with the occurrence of three major accidents in the Brazilian offshore oil and gas industry. This classification was chosen because it is based on the principles of HPI, used in the nuclear industry, and because it is applicable to studies that analyze human performance and investigating events in complex systems¹⁹,²⁰,²¹.

Methods

The classification proposed by Muschara¹⁹ was applied to three accidents considered of great relevance to the Brazilian offshore oil and gas industry : the explosion on the Floating Production Storage and Offloading (FPSO) Cidade de São Mateus, which occurred on February 11, 2015²²; the oil spill on the Petrobras 53 platform (P-53), recorded on March 24, 2019²³; and the accident involving the accidental release of carbon dioxide on the Petrobras platform (P-19), on August 2, 2022²⁴. The study used a qualitative and descriptive approach, based on documentary analysis of the official accident investigation reports issued by ANP²⁵. These reports are publicly accessible and are available online on the ANP website.

The choice of these events is justified by four main criteria: (i) the diversity of operational scenarios, involving an explosion, loss of containment and release of gas; (ii) the detail of the information in the official ANP reports, allowing for a structured analysis of the events; (iii) recent relevance for the Brazilian industry, since the accidents occurred in different periods and illustrate challenges present in the safety and integrity management systems of offshore installations; and (iv) significant impacts on the health and safety of workers.

The reports issued by the technical area responsible for operational safety at the ANP were used as the primary source of data for identifying the root causes associated with each event. The three reports analyzed used the fault tree methodology to define the root causes. There was no reinterpretation or revision of the root causes established by the ANP, which were adopted as an official reference and served as the basis for the complementary analysis of local factors developed in this study.

ANP Resolution No. 882/2022 defines a root cause as a failure in the management system that enabled the occurrence or existence of the causal factors of the accident under investigation⁷.

The analysis was conducted in two stages. In the first, the information contained in the official reports was technically read and categorized. In the second stage, the root causes were classified according to the six local factors defined by Muschara¹⁹: Requirements, Expectations, and Feedback; Tools and Resources; Incentives and Disincentives; Knowledge and Skills; Capacity; and Expectations, Motivation, and Personal Preferences.

Each root cause was assigned a local factor category according to the taxonomy used in the study. This classification was based on the original description of the causes in the ANP reports, to preserve the meaning attributed by the researchers and avoid inferences not supported by the official text.

The reports analyzed in this study were drawn up over a period of around ten years, considering different operational, regulatory and technical contexts. As official documents aimed at investigating complex events, these reports have varying levels of analytical depth and may contain methodological limitations inherent to the technical approach adopted in each case, as well as biases in the investigation process.

The classification of local factors proposed by Muschara¹⁹ was used as an analytical tool to contribute to the investigation practices adopted in the official ANP reports, and not as an exhaustive explanatory model of the accidents analyzed.

Results

The first stage of the work extracted from the final investigation reports the descriptions of the events and the root causes identified. Although these data do not constitute analytical results of this study, they represent elements of the basis used and are presented with the aim of contextualizing the accidents and preparing for the root cause classification stage. Basic information on the accidents is presented below.

The FPSO Cidade de São Mateus accident resulted from a major explosion in the unit’s pump room, during an operation related to the processing and transfer of hydrocarbons liquids of low density and viscosity. The event caused nine fatalities and 26 injuries, including ten workers in serious condition, as well as severe structural damage that led to the unit being permanently decommissioned. The ANP report identified failures related to the improper use of tanks, the absence of interlocks, equipment degradation and an insufficient change management process²². A total of 28 root causes were identified in this accident.

On the P-53 platform, the accident involved a leak of approximately 122 m³ of oil into the sea that hit the coast of the Lakes Region of the state of Rio de Janeiro, due to a fault in a line that had been operating for years in a temporary condition. The incident did not result in any injuries or fatalities but required an immediate environmental response with the activation of the emergency response plan for oil spills at sea. The investigation pointed to deficiencies in the change management process, prolonged maintenance of provisional arrangements, gaps in technical documentation and weaknesses in structural integrity control²³. Five root causes were identified in this accident.

The accident on platform P-19 occurred during a maintenance service in the auxiliary diesel motor generator room, when the carbon dioxide (CO₂ fire-fighting system was spuriously activated, leading to the death of one of the workers. The ANP identified shortcomings in the physical locking system, insufficient signaling, gaps in knowledge about CO₂ risks, as well as deficiencies in supervision and communication²⁴. Five root causes were identified in this accident.

In the second stage, 34 of the 38 root causes identified were classified as resulting from external factors, while only the other four root causes corresponded to internal factors, as shown in Table 1. This result is in line with the literature¹⁹ which shows that operational failures are mostly conditioned by the organizational context and project limitations, rather than by individual worker characteristics.

Table 1
Classification of Local Factors

Incident	Number of root causes per accident	Local factors
		External factors (N)			Internal factors (N)
		Requirements, expectations, and feedback	Tools and resources	Incentives and disincentives	Knowledge and skills	Capacity	Expectations, motivation, and personal preferences
FPSO Cidade de São Mateus	28	12	10	4	2	-	-
Platform P-53	5	3	1	-	1	-	-
Platform P-19	5	3	1	-	-	1	-

• FPSO: Floating Production Storage and Offloading

The breakdown by category (external and internal factors) showed that “Requirements, Expectations, and Feedback” was the most frequent external local factor, totaling 18 root causes, followed by “Tools and Resources”, with 12 records. Both factors are directly associated with the structural and organizational conditions that shape the execution of the work, showing that decisions, guidelines, and resources provided by the operators played a determining role in the evolution of accidents. The “Incentives and Disincentives” classification appeared in four root causes, reflecting situations in which production pressures and operational targets influenced the prioritization of operational continuity to the detriment of safety, a phenomenon widely documented in complex socio-technical systems¹⁹,²¹.

Internal local factors were less prevalent, totaling four records distributed between Knowledge and Skills (three cases) and Capacity (one case). Although less frequent, these internal factors appeared in critical situations mainly related to the P-19 platform accident, in which gaps in understanding about CO₂ risks were associated with behaviors that resulted in undue exposure to the contaminated atmosphere. The investigation report into the P-19 platform indicates that workers made relevant decisions without fully understanding the risks to which they were exposed. Nevertheless, even in these cases, internal factors were intertwined with external conditions, such as insufficient procedures, lack of signage and coordination failures.

Discussion

The distribution of local factors found in the accident analysis is close to the results presented by Muschara¹⁹ in their analysis of the nuclear industry, in which 75% of failures were related to external local factors and 25% to internal ones. Although the disproportion observed in this study is even more marked, the direction of the findings is convergent, since most failures are not the result of individual conduct, but of systemic conditions that influence and limit human performance. This alignment reinforces the applicability of Muschara’s classification¹⁹ in the context of the Brazilian offshore industry and indicates that adopting this approach can broaden the understanding of the events analyzed, by offering a structure that highlights the influence of organizational and individual factors in the official investigations conducted by the ANP.

The three accidents analyzed reinforced the structural predominance of external factors. On the FPSO Cidade de São Mateus, the root cause records showed significant failures in Requirements, Expectations, and Feedback and in Tools and Resources, mainly associated with change management, lack of alarms and insufficient protection devices. On the P-53 platform, the predominant causes were linked to the maintenance of provisional conditions, weaknesses in change management, and gaps in operational communication, revealing the action of external factors related to organizational incentives and insufficient rules and procedures. In the case of the P-19 platform, although internal factors were more evident, such as gaps in knowledge about CO₂ risks, it was also possible to identify external factors, such as the absence of physical blockades, supervisory failures, and planning deficiencies.

It should be noted that the analyzed cases, consisting of three accidents, preclude statistical analysis or quantitative generalization of the results. This limitation, however, is compatible with the qualitative design of the study, whose focus is not to measure frequencies, but to understand how the local factors defined by Muschara¹⁹ manifest themselves in different operational scenarios. Thus, the proportions presented act as descriptive indicators of the pattern observed in the cases analyzed and not as population metrics.

The results point to convergence between the cases and a widely discussed common narrative: human error does not appear as the primary cause, but as a consequence of accumulated systemic failures, a phenomenon already described by Reason²¹ as the dynamics of latent failures. Thus, the qualitative analysis emphasizes that accident prevention in the Brazilian offshore industry depends above all on strengthening SGSO management practices, especially those related to information management, change management, asset integrity, risk analysis and operational communication.

In summary, the classification of the 38 root causes based on Muschara’s local factors¹⁹ makes it possible to visualize the structuring role of external factors in the occurrence of accidents. This perspective reinforces the need for investigations that go beyond identifying human error and seek to understand the organizational and design conditions that shape operational behavior.

It is important to note that the classification of local factors proposed by Muschara¹⁹ has proved useful in reinterpreting the root causes identified in official ANP reports, offering a contribution to current investigation practices. The classification has a formal categorical structure and is oriented towards the analysis of performance factors, which can be interpreted as epistemologically distinct from the propositions of Safety-II and Resilience Engineering. However, in this study, its use was not intended to support a linear or reductionist logic of causality, but rather to structure the empirical analysis of local factors in a systematic and comparable way between the cases investigated. Additionally, it is recognized that the taxonomy, although effective in explaining organizational and individual influences, operates with limitations for the complete analysis of the event. Thus, it is understood that the application of the classification represents a contribution to structuring the analysis of human factors, by highlighting organizational and individual dimensions.

Conclusion

The analysis of the three accidents using Muschara’s classification¹⁹ showed that human error is a symptom of systemic failures consistent with highly complex socio-technical systems. In all the cases analyzed in this study, external local factors may have influenced human performance in highly complex contexts.

The results indicated that communication, insufficient feedback and inadequate organizational incentives played a predominant role. Muschara’s¹⁹ classification proved to be effective in connecting human behavior to organizational conditions, in line with the ANP and IOGP guidelines.

Systematically incorporating the analysis of local factors into investigations can strengthen organizational learning and improve the ANP’s SGSO.

The identification of external and internal factors shifts the focus of the investigation from individual error to the conditions that shape behavior, in line with the systemic approach to accidents proposed byReason²¹. By highlighting how human performance is a product of the interaction between individual capabilities and organizational and operational conditions, the approach reinforces the need to understand real work¹⁴ and the adaptations that workers make in complex systems.

It is recognized that improving accident investigations can involve various methodological fronts, such as in-depth analysis of barriers, operational changes and decision-making processes. However, from the perspective of human factors, based on the classification proposed by Muschara¹⁹, it stands out its potential to broaden the understanding of human performance in complex organizational contexts. As a direction for future research, we suggest further exploring the relationship between classification approaches and reference frameworks based on resilience and Safety-II, including their points of convergence and epistemological limitations.

References

• ¹ Hollnagel E. Safety-I and Safety-II: the past and future of safety management. Farnham: Ashgate; 2014.
• ² Brasil. Lei nº 9.478, de 6 de agosto de 1997. Dispõe sobre a política energética nacional, as atividades relativas ao monopólio do petróleo e institui o Conselho Nacional de Política Energética e a Agência Nacional do Petróleo e dá outras providências. Diário Oficial União. 7 ago 1997.
• ³ Ministério de Minas e Energia (BR). Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Boletim Mensal da Produção de Petróleo e Gás Natural. Set 2023. Rio de Janeiro: Ministério de Minas e Energia; 2023 [citado 10 out 2025]. Disponível em https://www.gov.br/anp/pt-br/centrais-de-conteudo/publicacoes/boletins-anp/boletim-mensal-da-producao-de-petroleo-e-gas-natural
  » https://www.gov.br/anp/pt-br/centrais-de-conteudo/publicacoes/boletins-anp/boletim-mensal-da-producao-de-petroleo-e-gas-natural
• ⁴ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Superintendência da Segurança Operacional. Relatório anual de segurança operacional das atividades de exploração e produção de petróleo e gás natural - 2023. Rio de Janeiro: Agência Nacional do Petróleo, Gás Natural e Biocombustíveis; 2024 [citado 10 out 2025]. Disponível em: https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/arq/raso/2023-relatorio-anual-seguranca-operacional.pdf
  » https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/arq/raso/2023-relatorio-anual-seguranca-operacional.pdf
• ⁵ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Resolução ANP nº 43, de 6 de dezembro de 2007. Estabelece o Regulamento Técnico do Sistema de Gerenciamento da Segurança Operacional (SGSO) de instalações marítimas de perfuração e produção de petróleo e gás natural. Diário Oficial União. 7 dez 2007.
• ⁶ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Resolução ANP nº 851, de 20 de setembro de 2021. Regulamenta o procedimento de fiscalização de segurança operacional. Diário Oficial da União. 2021. Disponível em: https://atosoficiais.com.br/anp/resolucao-n-851-2021
  » https://atosoficiais.com.br/anp/resolucao-n-851-2021
• ⁷ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Resolução ANP nº 882, de 27 de julho de 2022. Estabelece o procedimento para a comunicação de incidentes. Diário Oficial da União. 2022. Disponível em: https://atosoficiais.com.br/anp/resolucao-n-882-2022
  » https://atosoficiais.com.br/anp/resolucao-n-882-2022
• ⁸ Hollnagel E, Wears RL, Braithwaite J. From Safety-I to Safety-II: a white paper. Denmark: University of Southern Denmark; 2015.
• ⁹ Dekker S. The field guide to understanding human error. 3rd ed. Boca Raton: CRC Press; 2014.
• ¹⁰ Amalberti R. Navigating safety: necessary compromises and trade-offs. Heidelberg: Springer; 2013.
• ¹¹ Hollnagel E, Woods DD, Leveson N, editors. Resilience engineering: concepts and precepts. Aldershot: Ashgate; 2006.
• ¹² Frankenfeld KP, Mattos UA. organizadores. Engenharia de resiliência: fundamentos, aplicações e tendências. São Paulo: Letra Capital; 2021.
• ¹³ Chrusciak CB, Poncini CR, Moggio IH, Yasue JE, Bitencourt RS. Ergonomia e fatores humanos: um panorama das definições com base na literatura. Ação Ergonômica. 2020;14(1):62-74. https://doi.org/10.17648/rea.v14i1-12
  » https://doi.org/10.17648/rea.v14i1-12
• ¹⁴ Guérin F, Laville A, Daniellou F, Duraffourg J, Kerguelen A. Compreender o trabalho para transformá-lo: a prática da ergonomia. 2a ed. São Paulo: Edgard Blücher; 2011.
• ¹⁵ Daniellou F, Simard M, Boissière I. Human and organizational factors of safety: state of the art. Toulouse: FonCSI; 2011.
• ¹⁶ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Superintendência de Segurança Operacional. Coordenação de Segurança Operacional. Nota Técnica nº 10/2023/SSO/ANP-RJ: Disseminação de melhores práticas da indústria em Fatores Humanos. Rio de Janeiro: ANP; 2023. Disponível em: https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/arq/nota-tecnica-10-2023.pdf
  » https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/arq/nota-tecnica-10-2023.pdf
• ¹⁷ International Association of Oil & Gas Producers. Report 454: Human factors engineering in projects. London: IOGP; 2018 [citado 10 out 2025]. Disponível em: https://www.iogp.org/bookstore/product/human-factors-engineering-in-projects/
  » https://www.iogp.org/bookstore/product/human-factors-engineering-in-projects/
• ¹⁸ International Association of Oil & Gas Producers. Report 621: Learning from normal work and human performance. London: IOGP; 2020 [citado 10 out 2025]. Disponível em: https://www.iogp.org/bookstore/product/iogp-report-621prt-desmistificando-fatores-humanos-construindo-confianca-na-investigacao-de-fatores-humanos-demystifying-human-factors-building-confidence-in-human-factors-investigation-portugues/
  » https://www.iogp.org/bookstore/product/iogp-report-621prt-desmistificando-fatores-humanos-construindo-confianca-na-investigacao-de-fatores-humanos-demystifying-human-factors-building-confidence-in-human-factors-investigation-portugues/
• ¹⁹ Muschara TJ. Risk-based thinking: managing the uncertainty of human error in operations. New York: Routledge; 2018.
• ²⁰ Petrowski M, Warga J, Brown W, Gagan L, Baumann S, Newman J, et al. Best practice: the intersection of HPI and work planning and control. Los Alamos: National Laboratory; 2022.
• ²¹ Reason J. Managing the risks of organizational accidents. Aldershot: Ashgate; 1997.
• ²² Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Superintendência de Segurança Operacional e Meio Ambiente. Relatório de investigação do incidente de explosão ocorrido em 11/02/2015 no FPSO Cidade de São Mateus. Rio de Janeiro: Agência Nacional do Petróleo, Gás Natural e Biocombustíveis; 2015 [citado 16 nov 2025]. Disponível em: https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/incidentes/relatorios-de-investigacao-de-incidentes-1/arquivos-relatorios-de-investigacao-de-incidentes/fpso-cidade-de-sao-mateus/relatorio-de-investigacao-fpso-cidade-de-sao-mateus.pdf
  » https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/incidentes/relatorios-de-investigacao-de-incidentes-1/arquivos-relatorios-de-investigacao-de-incidentes/fpso-cidade-de-sao-mateus/relatorio-de-investigacao-fpso-cidade-de-sao-mateus.pdf
• ²³ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Superintendência de Segurança Operacional e Meio Ambiente. Relatório de investigação de incidente fpu-53 (vazamento de óleo com toque na costa). Rio de Janeiro: Agência Nacional do Petróleo, Gás Natural e Biocombustíveis; 2020 [citado 16 nov 2025]. Disponível em: https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/incidentes/relatorios-de-investigacao-de-incidentes-1/arquivos-relatorios-de-investigacao-de-incidentes/plataforma-p-53/relatorio-p-53_final.pdf
  » https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/incidentes/relatorios-de-investigacao-de-incidentes-1/arquivos-relatorios-de-investigacao-de-incidentes/plataforma-p-53/relatorio-p-53_final.pdf
• ²⁴ Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. Superintendência de Segurança Operacional. Relatório de investigação do incidente: Plataforma Petrobrás 19 (P-19). Rio de Janeiro: Agência Nacional do Petróleo, Gás Natural e Biocombustíveis; 2023 [citado 16 nov 2025]. Disponível em: https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/incidentes/relatorios-de-investigacao-de-incidentes-1/arquivos-relatorios-de-investigacao-de-incidentes/p-19/relatorio-investigacao-p-19.pdf
  » https://www.gov.br/anp/pt-br/assuntos/exploracao-e-producao-de-oleo-e-gas/seguranca-operacional/incidentes/relatorios-de-investigacao-de-incidentes-1/arquivos-relatorios-de-investigacao-de-incidentes/p-19/relatorio-investigacao-p-19.pdf
• ²⁵ Santos AR. Metodologia científica: a construção do conhecimento. 7a ed. rev. conforme BR14274:2005. Rio de Janeiro: Lamparina; 2007.

• Data availability:
  The entire data set supporting the results of this study is available on request from the corresponding author.
• Declaration on the use of Artificial Intelligence:
  The authors declare that no artificial intelligence tools were used in the preparation of the article.
• Presentation at a scientific event:
  The authors inform that the study has not been presented at a scientific event.
• Funding:
  The authors declare that the study was not subsidized.

Edited by

• Responsible editors:
  José Marçal Jackson Filho https://orcid.org/0000-0002-4944-5217 Fundação Jorge Duprat Figueiredo de Segurança e Medicina do Trabalho - FUNDACENTRO
  Raoni Rocha Simões https://orcid.org/0000-0003-1181-0132 Universidade Federal de Ouro Preto - UFOP

Data availability

The entire data set supporting the results of this study is available on request from the corresponding author.

Publication Dates

• Publication in this collection
  15 July 2026
• Date of issue
  2026

History

• Received
  13 Dec 2025
• Reviewed
  24 Jan 2026
• Accepted
  30 Jan 2026