Analysis of occupational accidents for safety design

Abstract

BACKGROUND:

Safety design covers proactive actions as it analyzes accident risks early in the enterprise life cycle, and considers the designer acting on accident prevention as a member of the construction team.

OBJECTIVE:

This paper proposes an accident investigation to establish links between accident causes and design to support Prevention through Design (PtD) tools.

METHODS:

This article analyzed more than a thousand severe and fatal accident cases in the construction sector. A systematic analysis method was structured based on descriptions of accident causes and measures that could be taken to avoid accidents.

RESULTS:

Analyzing the severe and fatal accidents, the safety measures implemented in the project design could avoid at least 23.6% of the events. As a result, the architectural and structural designs were more effective in accident prevention. The reference percentages and the design types that are more effective in preventing accidents are analyzed through a representative sample of the analysis of the accident.

CONCLUSIONS:

This research contributes to applying safety guidelines in design projects, directly assisting in project and construction management.

Keywords

Design for safety building safety risk identification

1 Introduction

Accident prevention in the design phase is characterized by proactive and efficacious actions, as it analyzes accident risks early in the project’s life cycle to ensure safety measures in the execution, maintenance, and deconstruction phases. This subject causes the involvement and interest of the entire construction chain.

The present work considered the definition of Prevention through Design (PtD) limited to permanent structure projects. Exemplifying, the concept does not focus on the use of fall protection systems but includes design decisions that influence how these protections will be needed. Similarly, that does not guide the involvement of scaffolding, but makes decisions that influence the location and type of scaffolding to perform the work [28]. The analysis of the construction project in its initial phase allows the scenarios’ description based on the visualization of possible events [27], a cost-effective way to avoid accidents, downtime, and improvisation in execution, maintenance, and deconstruction in construction sites. With the detection of the problems identified in the design phase, it is possible to establish preventive mechanisms and, apply measures to eliminate potential risks [25].

The methods that aim for risk prevention in the design and project planning phases and can contribute to several improvements for future construction. The efficient construction provides benefits for both the company and the employee, such as: increased productivity, which in turn directly affects the quality of the process and minimizes costs [36]. In general, the literature presents the benefits of accident prevention through design, besides the feasibility, models, and projections for the future. Design solutions already exist, but the challenge is to make changes to ensure that risks and hazards are eliminated and/or minimized [5].

The trained professional, in this case designers, was identified as a stakeholder that has an impact on occupational safety, although historically, designers have not considered safety in their designs and often do not know the the impacts of their design decisions on construction safety [32]. The PtD provokes the involvement and interest of the entire construction sector chain, as it is an economical way to avoid accidents, downtime and improvisation in execution, maintenance and deconstruction work. The choices of designers and the materials used are the main factors that determine the health and safety level of construction practices, and new methods are needed to determine the existing relationships between construction safety and designs [30].

The following hypotheses are raised in search of effectiveness and objectivity for design solutions that affect changes in construction site safety:

To understand how the project can contribute to the accidents prevention context it is necessary to know the hazards of accidents that could have reduced from the design;

To know these hazards it is necessary to know the accidents generated; and,

To know the accident it is necessary to know its causes.

Thus, there is a need to know the root causes of accidents and identify possible links with the designs, contributing to their reduction. Establishing the link between the causes of accidents and the designs in a representative sample, reference percentages can be generated, corroborating or refuting design-related accident percentages published and widely disseminated in the construction industry. In addition, there is a potential to direct solutions for prevention through design.

2 Literature review

Concerning the minimum safety and health requirements for temporary or mobile construction sites, the preamble of Council Directive 92/57/EEC of June 24th, 1992, found that inadequate architectural and/or organizational choices or even poor planning in project design contributed to more than half of the accidents at workplaces in the European Union [7]. In this sense, several studies and research have been performed to establish links between the causes of workplace accidents and designs.

In a study conducted in 1991, the European Foundation for the Improvement of Living and Working Conditions found that 60% of fatal accidents were a result of decisions made before the beginning of construction site activities. These accidents could have been avoided, by adopting measures in the design phase. Despite of the use this percentage in international articles (according to the articles referenced in this research), there is no published evidence to justify this rate.

In Australia, studies funded by the National Occupational Health & Safety Commission –NOHSC (now Safe Work Australia), providing statistical data on fatalities between July 2000 and June 2002, showed that 37% of the 210 fatal accidents analyzed were definitely/probably design-related [7].

Using an ergonomics approach, in 2003 researchers studied 100 construction accidents in the United Kingdom to identify where and why safety was compromised [17]. The findings stated that the designs could reduce risks in almost 50% of accidents.

Studies conducted in the Iranian construction industry showed that 33% of accidents were related to decisions made during the design phase [15].

In 2006, to determine whether safety considerations in design could prevent accidents, researchers evaluated 226 fatal accidents reports from Occupational Safety and Health Administration (OSHA), and 224 accidents reports from Fatality Assessment and Control Evaluation (FACE) all of them occurred in the United States [3]. The research used design safety suggestions [12] to discover a link between the accident causes and the designs. The research stipulated three types of answers: “yes”, “perhaps” and “no”. As a result, the author found that in 151 cases whose analyses indicated “yes” and “perhaps” answers, the risk that contributed to the accident could have been eliminated or reduced with the implementation of safety measures in the designs.

Therefore, establishing links between causes and designs through structuring methods is a worldwide concern and provides support tools for designer accident prevention. However, it is worth emphasizing the importance of validating the proposed methods through their applications with from a representative sample of occupational accidents. In this sense, this paper aims to present an accident analysis method and establish links between the causes of accidents and designs, using primary data with a sample of 1,328 cases of severe and fatal accidents, to generate reference percentages and guidelines for application in design projects.

3 Methods

Initially, the research retrieved work-related accidents on official sites utilizing work-related accident records. However, to compose the sample, the accidents were not predetermined, since this could compromise the results by the diversity of the factors involved in accidents, among them: type and phase of the work, construction methods, and systems, as well as the level of training of workers.

The proposed method used a sample composed of severe and fatal accidents from several countries and, therefore, with heterogeneous characteristics due to cultural, social, and economic diversity. The large sample of accidents highlighted the relevance of structuring a systematic analysis method to organize data, establish links between accidents and projects, and indicate project safety measures to prevent accidents. For this, this research studied existing methods of accident investigation. Some of these methods identified links between accident causes and designs. However, no information was found about the type of design that prevented the accident and the design measures.

The proposed method was based on descriptions of accident causes and measures to avoid accidents. Therefore, if the measures apply to the design, the analysis continues and the link between the accident causes and the design is analyzed. Then, the method requests the identification of the design types that could act in the accident prevention.

Finally, indicators are developed once the proposed occupational accident method is structured and applied. Therefore, it translates the analysis results into percentages to use as reference values on the participation of the designs in the causes of workplace accidents.

The description of the research methodology is divided into three parts:

Collection of work accidents in the construction industry;

Structuring the proposed method for the analysis of work accidents;

Workplace accident analysis to establish any links between accident causes and designs/Development of indicators.

3.1 Collection of construction work accidents

The collection of work accidents was made by searching for accident reports in governmental and private institutions’ databases worldwide. However, most virtual databases of work accidents cite only immediate causes (such as falls, electric shock, burial, etc.) and do not describe the accident at work. Therefore, only the cases with the respective accident descriptions in their content, considered fundamental to the analysis performed below, were collected.

The research collected reports of work-related accidents of Brazil, Canada, Portugal, the United States, and Singapore. Table 1 presents the quantitative data of the accident cases collected and analyzed, by country, with the respective references, periods of occurrence, and nature of the accident. The data were analyzed and treated individually due to cultural, social, and economic diversity between countries. Each country’s context reflects differently the workers’ training, construction mechanization, construction methods, and systems, among other aspects. It is worth noting the differences in design practices and project regulations between the countries, but which are not relevant in this study since the analysis is focused on accidents that have already occurred, and there was no intention to evaluate the respective designs.

Table 1
Accidents collected and analyzed in the construction industry: partial figures by country

Country Reference Number of occupational accidents Period Accident nature

Brazil/Pernambuco SFIT apud Maia (2008) 32 2002–2006 Fatal

Canada CCOHS 940 1977–2008 Fatal

United States FACE/NIOSH –PtD 112 1991–2011 Fatal

Portugal ACT apud Reis (2005) 203 2000 Serious and fatal

Singapore WSH Concil 41 Non-informed Fatal

Total 1328

Country	Reference	Number of occupational accidents	Period	Accident nature
Brazil/Pernambuco	SFIT apud Maia (2008)	32	2002–2006	Fatal
Canada	CCOHS	940	1977–2008	Fatal
United States	FACE/NIOSH –PtD	112	1991–2011	Fatal
Portugal	ACT apud Reis (2005)	203	2000	Serious and fatal
Singapore	WSH Concil	41	Non-informed	Fatal
Total		1328

Source: The authors.

3.2 Structuring the proposed method for analysis of occupational accidents

Initially, the research investigated the existing methods related to the influence of designs on the causes of work-related accidents. As mentioned, some studies have published percentages associated with the designs’ participation, ranging from 22% to 60% [1 , 17].

The proposed method relies on descriptions of the causes of accidents and the associated measures to prevent accidents. Thus, an analysis is performed by reading these fields in the reports. It is relevant to highlight that these fields are usual in the documents examined, regardless of the country. In addition, the local authorities are responsible for filing the reports of accident investigation and there is no space for interpretation and/or subjectivity of who is using the method.

Then, based on these measures to prevent accidents, the method analyzes an answer and the link between accident causes and designs, providing the options “yes” and “no”. If the answer is yes, the method provides a space for answers about the types of designs involved and the possible safety guidelines for design. Intentionally, the method is dichotomous because terms like “maybe”, “probably” or “possibly” open gaps for subjectivity or doubt. The binary mode was chosen since the method intends to find the percentage of accidents avoided.

By linking the accident to designs, it indicates the choice of the design type. Figure 1 shows the classification and characterization of the designs through the method application.

Fig. 1

Design category, type and characterization.

Figure 1 shows three categories of designs: design projects, execution designs, and equipment designs, associated with the types and their characteristics. Design projects refer to designs for permanent structures. Thus, they provide a basis for other disciplines subdivided into architectural, structural, and facility designs. The latter covers various types of facilities.

The execution design supports the execution of the work. This category involves designs related to the construction site, such as signage, temporary facilities, the definition of access, circulation, cargo handling and storage of construction materials, and safety design. The safety designs differ in countries and permeate transversally in all project phases. In this case, it is considered specific designs in the execution phase of the work, such as scaffolding, lifelines, platforms, etc. Equipment design refers to the design of equipment, machines, and tools used in the execution of the enterprise (being classified and presented in Fig. 1).

Considering the classification of projects contained in Fig. 1, it was verified that for the proposed method, the type of work should also be considered in the classification of projects. Figure 2 classifies the works into two groups: building and infrastructure. The latter is also subdivided into two types: civil engineering/heavy construction and industrial assembly. In the last column, the table shows the characterization of each type of work. In the infrastructure, structure and installations designs were compiled because, generally, infrastructure designers are responsible for both natures.

Fig. 2

Construction work type and characterization.

The types of design identification linked to the accident lead to the choice of adequate guidelines in the design projects, available in a list of the method itself. To structure this list, design guidelines/recommendations/suggestions available in the literature were collected, analyzed and compiled, totaling 1061 items [1 , 34]. However, a large part of these guidelines/recommendations/suggestions would not fit the analysis of the collected accidents because many propositions are repetitive and/or refer to designs that are not part of the project designs. Therefore, the list of the proposed method considers only guidelines directed to architectural, structural, installation, and infrastructure designs.

Although easy to use, the method has limitations because it relies on previous technical knowledge in construction safety by those who apply it. In this sense, it is recommended that the person applying it be an architect or civil engineer with technical knowledge in occupational safety. On the other hand, the method was validated through a case study developed by collecting and analyzing the accidents that occurred at a construction site. This construction site developed activities in a constructed area of 295,000 m², besides works related to two viaducts, riverbank revitalization, and landscaping, having a peak of 6924 employees and a sum of 409 service providing companies. Given the size of the construction site, the diversity and complexity of the activities and functions involved in its execution, and, consequently, the diversity of the associated accident risks, the choice of this construction site for the case study was deemed appropriate. Another relevant factor for the choice was that it was a building construction work. The case study data can be made available for consultation, through contact with the authors.

3.3 Analysis of occupational accidents in the construction industry and indicator structuring

The method was applied initially by an architect with a background in construction safety, and there were weekly systematic discussions with two civil engineers, both with expertise in construction safety. In addition, the present work is the result of a doctoral thesis that was submitted and approved with distinction by a panel of five professors from different universities.

Firstly, the accidents were analyzed separately by country due to the diversity of factors involved in the accidents, such as the type and stage of the work, the construction methods and systems, and the level of training of the workers. After applying the method, the data were processed, generating the indicators of Fig. 3:

Fig. 3

Indicators.

4 Results

The results are divided into three parts. The first part refers to the proposed method. The second part considers the analysis of accidents and indicators (general and by design category). Finally, the third part presents the indicators by design type.

4.1 Proposed method for analysis of occupational accidents

The proposed method to analyze the work-related accidents collected, called MAARD –Method of Analysis for Accident-Related Design –MAARD, is a systematic analysis method composed of two integrated worksheets. Figure 4 shows the MAARD scheme.

Fig. 4

AARD scheme.

The first worksheet (Fig. 5) Descriptive Information, is comprised of the following fields:

Occupation of the injured worker: This field identifies the occupation of the injured worker.

The severity of the accident: Minor, serious, and fatal options are available, limited to the information contained in accident reports.

Cause of the accident: For this field, sometimes the accident had to be rewritten and/or summarized, taking care to preserve the content since some reports presented repetitions and informal language. The accident is excluded if there is not enough information to analyze the causes.

Fig. 5

Application example MAARD –Worksheet 1 (is summarized).

Fig. 6

Application example MAARD –Worksheet 2 (is summarized).

The second worksheet (Fig. 6) Analytical Information is comprised of the following fields:

Causal Factors: Although the work accident is multicausal, we sought to identify the most relevant factor for its occurrence, referring to the decisive and immediate aspect of the accident. Since the causal factors of an accident are related to technical safety failures, the following classification was adopted [3]: human factor, material factor, and organizational factor. The human factor refers to the workers’ actions as the most relevant factor for the occurrence of the accident. It is significant to highlight that it is not the intention of this work to identify the blame or responsibility for the accident, knowing that other factors can be involved in the worker’s decision-making. The material factor is limited to factors directly related to the inadequacies and/or failures of machines, equipment, and tools. On the other hand, the organizational factor involves actions of a more comprehensive dimension, in which the organizational policy and culture are directly or indirectly related. The lack and/or inadequacies and/or failures in construction site management, operational procedures, and/or projects are linked to this field.

Measures that could have been taken: In this field, the measures that could be taken so that the accident did not occur are described, focusing on the most relevant causal factor.

Could the design have helped to avoid the accident? The information contained in the previous field is analyzed and compared with the list of the guidelines for construction accident risk prevention for project design, presented in Table 7. Through this analysis, the answer is affirmative if any measures can be implemented in the design. This field requires direct yes/no answers to identify whether the design can prevent the accident from occurring.

In which design category? The design category (design project, execution design, and equipment design) that could have collaborated to avoid the accident is selected.

In which design type? In this field there is the selection of the type of design (architecture, structure, facilities, and infrastructure) that the designer could have collaborated to avoid the accident.

Accident risk prevention guidelines that could be adopted in the design project: This field is only for accidents involving design projects. The number of guidelines for construction accident risk prevention corresponding to the item from the list of construction accident prevention guidelines for the design project (Fig. 7) is indicated.

Fig. 7

List of construction accident risk prevention guideline for project designs.

As described, the list of guidelines took as reference 1061 compiled guidelines/recommendations/suggestions, which may be available. It is worth highlighting that structuring the guidelines list in Fig. 7 sought to preserve the designers’ freedom of creation with flexible guidelines. In some cases, the guidelines have not been used in an incisive way since the MAARD criteria do not condition the designer’s definitions, giving the professional freedom to make choices. Furthermore, new guidelines were created for the MAARD list that were not predicted by previous researches but originated during the accident analysis.

4.2 Analysis of accidents: general and design indicators

The accident analyses were organized by country, using the MAARD –Method of Analysis for Accident Related Design. Then, design influence indicators were prepared for each country, as shown in Table 2. Although extensive, the content of the accident analysis can be made available, with MAARD worksheets 1 and 2 filled in with the data of the 1,328 accidents.

Table 2
Participation of design (by category) in the causes of the accidents analyzed –by country

Country Overall Design category (%)

(%) Project Execution Equipment

design

Brazil (PE) 75,9 27,6 41,4 13,8

Canada 46,6 23,6 18,4 5,0

United States 49,1 40,7 22,8 23,6

Portugal 67,5 38,4 28,5 3,3

Singapore 65,0 45,0 25,0 2,5

Country	Overall	Design category (%)
Brazil (PE)	75,9	27,6	41,4	13,8
Canada	46,6	23,6	18,4	5,0
United States	49,1	40,7	22,8	23,6
Portugal	67,5	38,4	28,5	3,3
Singapore	65,0	45,0	25,0	2,5

Source: The authors.

Brazil

Regarding the State of Pernambuco/Brazil, 32 cases of fatal accidents from 2002 to 2006 were collected and analyzed. The analysis is based on a study [20] without any modification or synthesis of the reports. When analyzing the whole sample of Brazilian accidents, 75.9% of the cases could be avoided if preventive measures were implemented in the design. This percentage of accidents linked to the design is related to the immediate causes with the highest number of repetitions in accidents, such as falls from height and electric shock.

Considering the design categories (design/execution/equipment), several accidents by falling from height could have been avoided using preventive measures implemented in the designs, such as architectural and/or structural designs. In the cases of electrical shock accidents, some could have been avoided by utilizing measures implemented in the execution designs. Regarding the design category, the participation of the project designs represented almost a third (27.6%) of the accidents analyzed. Regarding execution and equipment designs, these were 41.4% and 13.8%, respectively.

The analysis of the data shows that the construction sites of the accidents analyzed did not yet have planned conditions and adequate execution designs (e.g. scaffolding project, lifeline project, access, circulation, cargo handling, material storage). Accident data also indicate the use of old equipment without periodic maintenance and without achieving the work safety requirements.

Canada

In Canada, 940 fatal accident cases from 1987 to 2008 were collected and analyzed. The accident descriptions provided information about the accident and its causes from the accident reports available in the Canadian Centre for Occupational Health and Safety (CCOHS) database [4].

A filter was applied in the database to compile only construction work accidents. During the MAARD application, accident reports with unavailable accident description fields were excluded.

Most of the accidents with causes linked to the project design were caused by falling from a height, followed by burial, electric shock, and machinery rollover. Another causal factor that was also significant in the Canadian accident cases was the human factor, as many accidents occurred due to the lack of workers’ training. It is worth highlighting that most of the injured workers were foreign. Therefore, some accidents reports analysed reinforced the need to promote training and write operating procedures in several languages as safety measures

In contrast, 46.6% of the Canadian accidents could have been avoided with preventive measures implemented in designs. Regarding the design categories related to work accidents, the design projects could have prevented 23.6% of the accidents analyzed i.e., the architectural, structural, facility and infrastructure designs.

In the case of execution and equipment designs, the percentages were 18.4% and 5%, respectively. The data shows that the construction sites of the accidents analyzed probably used machines and equipment with the necessary safety devices, besides establishing, in general, planned and adequate conditions for the execution designs. It is noteworthy that most of the accidents related to equipment designs refer to accidents that occurred in the 1980s and early 1990s. Between the 80s and 90s, there was no safety devices used in construction equipment as today, such as the reverse sensor. The Canadian Occupational Health and Safety Regulations SOR/96-400 require such a device. Canadian regulations are cited by SOR (Statutory Orders and Regulations) by year and regulation number, which means rear sensors in equipment, became mandatory in 1996. Therefore, in cases before this year, the rear sensor was identified as a measure that could avoid an accident, and the design of the equipment was linked to the causes of the accidents. On the other hand, in more recent accidents, the causes of the accidents point to other reasons, such as the failure of the reverse sensor due to lack of maintenance.

United States of America

As for accidents in the United States, 2,449 work-related accidents were found on the Centers for Disease Control and Prevention (CDC) website [24]. Through the Fatality Assessment and Control (FACE) program, the CDC provides research reports from the National Institute for Occupational Safety and Health (NIOSH). The total number of accident reports found refers to accidents that occurred in the various industrial sectors of the United States from 1982 to 2012. Among them, 722 were in the construction industry. The accident cases are divided into two groups. The first is the national reports: the NIOSH reports, with a total of 610 reports (247 within the construction industry). The second is the state-based reports: the state FACE reports, with a total of 1,839 accidents reports (475 in the construction industry).

The PtD access icon is used when accessing and downloading the files. This icon acts as a filter to access accident reports with relevant design recommendations, this filter was used only in the US sample. Thus, 455 fatal accident cases with design recommendations were identified after applying the filter. However, these accidents came from several industrial sectors, and it was necessary to perform a manual sorting to select only the accidents that occurred in civil construction. One hundred and twelve construction accident reports were recorded for the sample, and subsequently the MAARD application, as presented in Table 3.

Table 3

Constitution of the quantitative sample of accidents for analysis –United States

	Construction	Other
	industry	industrial
		sectors
NIOSH FACE PtD Reports	31	64
NIOSH State FACE PtD Reports	81	279
Subtotals	112	343
Total		455

Source: The authors.

Most accidents with the organizational factor as the most common causal factor refer to falls from height with many of them being design-related. Accidents involving machinery and equipment also represented a reasonable number of the total accidents analyzed (23.6%), and included the organizational and human factors as the predominant causal factors of the accident. Despite reinforcing the need for workers’ training, the cases of accidents involving equipment, machines, and tools in the USA reports, tend to automate and utilize technology to avoid workers’ decision-making.

In 49.1% of the cases, the accidents could have been avoided if preventive measures had been implemented in the designs. Regarding the design categories related to work accidents, in 40.7% of the accidents analyzed, the project designs could have prevented accidents. Regarding the execution and equipment designs, the percentages were 22.8% and 23.6%, respectively.

The data show that the construction sites of the accidents analyzed probably established adequate conditions for execution designs. The percentage of accidents related to equipment designs reflects the deep thinking of the competent American agencies in the development and use of new technologies in machine and equipment designs. Thus, it will contribute to increasing the mechanization level of activities and, consequently, reduce the need for on-the-job decision-making by the employee.

Portugal

Regarding Portuguese accidents, 203 severe and fatal work accidents occurred in 2000 [26]. In 67.5% of the cases, accidents could have been avoided if preventive measures had been implemented in the designs. Most Portuguese design-related accidents refer to falling from a height, which could be avoided in the design phase. Regarding the burial/demolition accidents, in 38.4% of the events analyzed project designs could have prevented the accidents.

In the case of execution and equipment designs, the percentages were 28.5% and 3.3%, respectively. The data show that the construction sites lacked adequate temporary planning and designs, such as the scaffolding, lifelines, loads, and storage of materials. On the other hand, the data indicate that Portuguese construction sites use machinery and equipment with the necessary safety devices.

Singapore

Regarding work accidents in Singapore, case studies were collected from the Workplace Safety and Health Council –WSH Council, available on the institutional website [37]. Thirty-one fatal accidents, were analyzed with specific cases of accidents due to falling from heights (23 accidents), falling objects/materials on workers (13), electrical shock (2), and mechanical shock against objects and machines (3). [35].

In the analysis of 41 accident cases in Singapore, 65% of the events could have been avoided if preventive measures had been implemented in the designs. More than half refer to accidents caused by falls from height. On the other hand, accidents caused by falls from height and accidents caused by falling objects and materials had their causes related to the designs in more than half of the cases. These data indicate that designs play a relevant role in eliminating and/or reducing the risks of falls from height and structure collapse, such as placing lifeline hooks in structural designs, using unbreakable glass or waiting on the facades for scales.

The contribution of project designs as a causal factor for accident represented almost half (45%) of the accidents analyzed. In absolute numbers, 18 fatal accidents could have been avoided with preventive measures in the project designs. The percentage contributions of execution and equipment designs issues were 25% and 2.5%, respectively. From the analysis of the data, the construction sites had neither adequate planning nor temporary designs, whether safety design scaffolding design, lifeline design, loads, storage of materials, etc. Only the machinery and equipment had the necessary safety devices.

4.3 Analysis of accidents –indicators by design type

For each country, indicators about the percentage contribution of the designs (by type) in the causes of the accidents analyzed were compiled in Table 4. It is highlighted that the link with the designs can occur in more than one type of design simultaneously.

Table 4
Representation of designs (by type) in accidents related to project designs-by country

Country Design type (%)

Architecture Structure Facilities Infrastructure

Brazil (PE) 62,5 75,0 12,5 12,5

Canada 52,6 60,8 5,8 26,3

United States 50,0 50,0 4,5 40,9

Portugal 72,4 75,9 6,9 13,8

Singapore 66,7 94,4 5,6 5,6

Country	Design type (%)
Brazil (PE)	62,5	75,0	12,5	12,5
Canada	52,6	60,8	5,8	26,3
United States	50,0	50,0	4,5	40,9
Portugal	72,4	75,9	6,9	13,8
Singapore	66,7	94,4	5,6	5,6

Source: The authors.

In the case of Brazil, architectural and structural designs had a majority contribution to the accidents, with 62.5% and 75%, respectively. As for the implementation of safety measures in installation and infrastructure designs they could have prevented only 12.5% of accidents each. Architecture and structure designs had high representation in Canadian accidents linked whith design, with 52.6% and 60.8%, respectively.. Facility and infrastructure design also played a role in accident prevention, with 5.8% and 26.3%, respectively. Regarding accidents in the USA, architecture and structure designs accounted for 11.50%, facility design for 4.5%, and infrastructure designs for 40.9% of fatal accidents. As in the results of Brazil, Canada, and the United States, the architectural and structural designs had high representation factors in the Portuguese accidents, with 72.4% and 75.9%, respectively. The implementation of safety measures in facility and infrastructure designs could have prevented respectively 6.9% and 13.8% of the accidents. Finally, in the accidents in Singapore, architecture and structure designs had high representation in accidents linked with design, with 66.7% and 94.4%,; respectively. The implementation of safety measures in facilities and infrastructure designs would have each prevented 5.6% of accidents.

Table 5 presents the indicators summary by country from the analysis of the accidents; the last three rows show the minimum, maximum, and average values of accidents with causes linked to design. The accidents data were analyzed separately, by country. However, averaging across all countries with data analyzed, the implementation of design measures could avoid an average of 60.8% of accidents among countries with severe and fatal events. In addition, utilizing the lowest percentage encountered, at least 46.6% of occupational accidents could have been prevented by safety measures implemented in project design, execution design, or equipment design.

Table 5

Designs (by category and by type) linked with accident causes (percentage representation), and analyzed by country

Country	Overall (%)	Design category (%)
		Design	Execution	Equipment
Brazil (PE)	75,9	27,6	41,4	13,8
Canada	46,6	23,6	18,4	5,0
United States	49,1	40,7	22,8	23,6
Portugal	67,5	38,4	28,5	3,3
Singapore	65,0	45,0	25,0	2,5
Minimum	46,6	23,6	18,4	2,5
Maximum	75,9	45,0	41,4	23,6
Average	60,8	35,1	27,2	9,6
Country	Type of design (%)
	Architecture	Structure	Facilities	Infrastructure
Brazil (PE)	62,5	75,0	12,5	12,5
Canada	52,6	60,8	5,8	26,3
United States	50,0	50,0	4,5	40,9
Portugal	72,4	75,9	6,9	13,8
Singapore	66,7	94,4	5,6	5,6
Minimum	50,0	50,0	4,5	5,6
Maximum	72,4	94,4	12,5	40,9
Average	60,8	71,2	7,1	19,8

Source: The authors.

The percentages are lower in severe and fatal accident reports with causes linked to project design, ranging from 23.6% to 45%. Thus, the implementation of preventive measures in the project designs, specifically in architecture, structure, facilities, and infrastructure designs, could avoid an average of 35.1% of the accidents. Table 5 also shows the representativeness of the design in accidents with project-related causes. Structure design was the most effective design for accident prevention, followed by the architecture design in all the countries that had occupational accidents analyzed in the period indicated in Table 1.

Structure design contributed to an average of 71.2% of causes of fatal accidents, with minimum and maximum values of 50% and 94.4%, respectively. The percentages of architectural design causal factors in fatal accidents were similar to those presented by structural design factors; with an average of 60.8% and minimum and maximum values of 50% and 72.4%, respectively.

The infrastructure design factors contributed to an average of 19.8% of causal factors, ranging from 5.6% and 40.9%, respectively. Facility design was the least representative causal factor for fatal accidents, with an average of 7.1%. The minimum and maximum values were 4.5% and 12.5%, respectively.

5 Discussion

In Brazil (Pernambuco), based on accidents reports, some accidents occurred in the construction phase, using power tools, machinery, and equipment. In general, construction sites in Pernambuco/Brazil lack planning and, consequently, there is improvisation in activities and installations. This fact explains the high number of cases with execution designs that could have avoided accidents (41.4%). Another observation about Brazilian accidents is that safety measures implemented in the equipment designs could have avoided around 13.8% of the events. Based on information from accidents reports, this percentage (13.8%) could indicate that the machinery and equipment used in the Brazilian construction sites analyzed are old, without the adoption of the more recent occupational safety requirements. In addition, there is a lack of preventive maintenance, only considering corrective maintenance after the occurrence of damage, due to the fast pace of the works.

The accident data in Canada and Singapore indicated that many accidents occurred in maintenance services with causes related to the project designs. As most accidents are due to falls from height as the immediate cause, the architectural and structural designs had greater representation in severe and fatal accidents. The descriptions of the accidents and the data treated from the analysis of the accidents in Canada and Singapore point to a high level of mechanization in the executive process at the construction sites, making use of machines and equipment consistent with the aspects of safety at work. The Canadian work accidents in the 1980s and early 1990s are exceptions when there were technological limitations in machinery and equipment design.

In the United States, accident reports involving machinery and equipment direct safety actions at work for manufacturers, in order to develop technologies that increase the level of safety and avoid decision-making in the workplace by operators. At the same time, the reports always reinforce the need for qualification and training of workers in different languages, which is also needed in Canada.

In Portugal, most accidents occurred when the structure phase is completed, which explains the proportionality between the indicators of accidents with causes linked to project designs (38.4%) and execution designs (28.5%). Another consideration is the recent introduction of new machinery and equipment on the construction sites that comply with current safety regulations. In addition, the accident reports showed that there is a lack of planning both at the construction sites and in workers’ training.

Finally, knowing that the elimination of hazards and risk reduction on construction sites is a complex issue, the PtD does not guarantee that risks will reduce or hazards will be eliminated. For example, the prevention of a fall accident can happen with a lifeline anchor on the deck of a building depicted in the architectural and structural designs if a series of consecutive safety measures happen. There is also a need to implement the correct installation of the anchorage and the training of the workers on the risk of falling and the use of a safety belt. In addition, workers must reproduce safe behavior, and it must be supervised by the employer. In other words, the PtD does not contribute to the predictability of accidents because there are factors of human behavior in subsequent stages, but it can collaborate proactively in accident prevention.

6 Conclusions

The causes of accidents at work can be numerous, arising from lack or deficiency of planning and production organization, unsafe conditions in the work environments, and human factors. However, whatever the causes, they can be detected early by the organization’s technical managers, and thus there is a need for action in the conception phase.

This study analyzed more than a thousand accidents that occurred in the State of Pernambuco/Brazil, Canada, the United States, Portugal, and Singapore. The data were treated separately by country due to the cultural, social, and economic diversity, which impacts worker training, construction mechanization, construction methods. On the other hand, the data can be available, providing transparency and information reliability. In addition, the availability can allow other researchers to use the same database for future comparisons.

The MAARD –Method of Analysis for Accident Related Design analysis method used the content of the work accident reports and accident risk prevention guidelines to establish the link between the causes of accidents and the designs. These guidelines are based on the choice, analysis, and treatment of more than a thousand safety items for designs published worldwide. Despite being easy to apply, the method has limitations because it depends on the technical knowledge of the person who applies it. In this sense, it is recommended that the profissional be an architect or civil engineer with technical knowledge in work safety.

Through the analysis of the collected accidents, the present work established relationships between accidents in construction and the projects to indicate a reference percentage and provide more reliability about the percentages published worldwide. Moreover, it specifies the types of projects that are more capable of preventing accidents. Consequently, it identifies the designers with a greater predisposition to collaborate effectively in accident prevention, justifying the creation of normative requirements for certain types of projects. The validation of the method will be presented later.

Ethical approval

Not applicable.

Informed consent

Not applicable.

Conflict of interest

Not applicable.

Data availability

This article summarizes the research data. However, other data can be made available for consultation if necessary.

Funding

Own resources.

Footnotes

Acknowledgments

The authors would like to thank the Occupational Safety and Hygiene Laboratory –LSHT for its facilities and support.

In memory of Professor Dr Béda Barkokébas Junior.

References

ASCC. [homepage on the Internet]. Australia; 2006. Commonwealth of Australia, Canberra. [updated 2012 July 01 cited 2012 July]. Available from: http://www.safeworkaustralia.gov.au/sites/SWA/about/Publications/Documents/154/GuidanceOnThePrinciplesOfSafeDesign_2006_PDF.pdf.

Behm

. Linking construction fatalities to the design for construction safety concept, Safety Science. Greenville. 2005;43:589–611.

Behm

[homepage on the Internet]. Greenville; 2006. An analysis of construction accidents from a design perspective. CPWR –The Center to Protect Workers’ Rights, Silver Spring. [updated 2012 July 01 cited 2012 July]. Available from:http://www.cpwr.com/rp-cpwrreports_design_for_safety.php.

Bellovi

, et al. Seguridade nel Trabajo. Instituto Nacional de Seguridad e Higiene enel trabajo, Barcelona; 1990.

Canadian Centre for Occupational Health and Safety –CCOHS. [homepage on the Internet]. Canada, 2009. [updated 2012 March 23 cited 2012 March].Available from: http://ccinfoweb.ccohs.ca/fatality/search.html.

Creaser

. Prevention Through Design (ptd): Safe Design from an Australian Perspective, Journal of Safety Research. Australia. 2008;39:131–4.

Driscoll

, Harrisson

, Bradley

, Newson

. The Role of Design Issues in Work-Related Fatal Injury in Australia, Journal of Safety Research. 2008;39:209–14.

EFCA, ACE [homepage on the Internet]. Brussels; 2006. Designing for safety in construction: Taking account of the ‘general principles of prevention. [updated 2012 Juny 01 cited 2012 July]. Available from: http://www.efcanet.org/Portals/EFCA/ELOKET/1232/DesigningForSafety.pdf.

EUR-Lex [homepage on the Internet]. Brasil; 1992. Diretiva 92/57/CEE do Conselho, de 24 de junho de 1992. [updated 2011 April 14 cited 2011 April]. Available from: http://europa.eu.in/eur-lex/.

10.

European Foundation for the Improvement of Living and Working Conditions. From drawing board to building site (EF/88/17/FR). European Foundation for the Improvement of Living and Working Conditions. Dublin; 1991.

11.

Fatality Assessment and Control Evaluation –FACE program [homepage on the Internet]. USA; 2008. [updated 2011 May 08 cited 2011 May]. Available from:http://www.cdc.gov/niosh/face/inhouse.html.

12.

Gambatese

, Behm

, Hinze

. Viability of designing for construction worker safety, Journal of Construction Engineering and Management.. 2005;131(9):1029–36.

13.

Gambatese

, Hinze

, Haas

. Tool to design for construction worker safety, Journal of Architectural Engineering. 1997;3(1):32–41.

14.

Gambatese

, Hinze

. Addressing construction worker safety in the design phase: Designing for construction worker safety, Automation in Construction. 1999;8(6):643–9.

15.

Gambatese

, Hinze

, Behm

[homepage on the Internet]. Seatle; 2005. Investigation of the viability of designing for safety, the center to protect worker’s safety, silver spring. [updated 2011 April 14 cited 2011 April]. Available from:http://www.cpwr.com/pdfs/pubs/research_pubs/krgambatese.pdf.

16.

Ghaderi

, Kasirossafar

. Construction safety in design process, Proceedings of the AEI Conference, ASCE Conf. Proc. California. 2011;2011(2).

17.

Hecker

, Gambatese

, Weinstein

. Designing for worker safety, Professional Safety. 2005;50:32–44.

18.

Hide

, Atkinson

, Pavitt

, Haslam

, Gibb

, Gyi

, Duff

, Suraji

Causal factors in construction accidents. HSE Books. Suffolk. Loughborough; 2003.

19.

Hinze

, Gambatese

Adressing construction worker safety in the project design. The University of Washington. Seattle; 1996.

20.

Hinze

, Wiegand

. Role of designers in construction worker safety, Journal of Construction Engineering and Management. 1992118(4).

21.

Maia

Análise de acidentes fatais na indústria da construção civil do Estado de Pernambuco. Master’s Thesis. Catholic University of Pernambuco. Recife; 2008. p. 122.Análise de acidentes fatais na indústria da construção civil do Estado de Pernambuco, Master’s Thesis. Catholic University of Pernambuco. Recife;. 2008122.

22.

Mroszczyc

[homepage on the Internet]. Designing for Construction Worker Safety M. [updated 2011 May 29 cited 2011 May]. Park Ridge; 2006. Available from: www.asse.org/practicespecialities/.../John-Mroszcyck_article.doc.

23.

NOHSC [homepage on the Internet].Work-related fatalities associate with design issues involving machinery and fixed plant in Australia, 1989 to 1992. [updated 2011May14 cited 2011 May]. NOHSC. Sydney; 2000. Available from: http://www.safeworkaustralia.gov.au/sites/swa/about/publicatio-ns/pages/sr2000fatalitieswithdesignissuesofmachinery1989to1992.

24.

NIOSH [homepage on the Internet]. Fatality Assessment and Control Evaluation (FACE) Program. [updated 2011 February 14 cited 2011 February]. USA; 2013. Available https://www.cdc.gov/niosh/face/inhouse.html.

25.

Raviv

, Shapira

, Sacks

. Empirical investigation of the applicability of constructability methods to prevent design errors, Built Environment Project and Asset Management. 2022;12(1):53–69.

26.

Reis

Melhoria da eficácia dos planos de segurança na redução de acidentes de construção, Doctoral Dissertation. Porto University. Porto;2008.

27.

Ruiz-Zafra

, Benghazi

, Noguera

. IFC+: Towards the integration of Io Tinto early stages of building design, Automation in Construction. 2022;136:1–17.

28.

Saurin

. Segurança do trabalho e desenvolvimento de produto: diretrizes para integração na construção civil, Revista Produção. 2005;5(1):127–41.

29.

Toole

, Hervol

, Hallowel

. Designing for construction safety, Modern Steel Construction. USA. 2006;46:55–9.

30.

Toole

, Hervol

, Hallowell

Design steel for construction safety. North American Steel Construction Conference, San Antonio. 2007.

31.

Torghabehe

, Hosseinian

. Designing for construction worker’s safety, International Journal of Advances in Engineering & Technology. 2012;4(2):373–82.

32.

Vasconcelos

Segurança do trabalho no projeto de arquitetura: diretrizes para o controle dos riscos de acidente na fase pós-obra. Master’s Thesis. Pernambuco University. Recife; 2009. p. 128.

33.

Vasconcelos

Segurança no trabalho na construção: modelo de gestão de prevenção de acidentes para a fase de concepção. Doctoral Thesis. Faculty of Engineering of University of Porto. Porto; 2013. p. 934.

34.

Weinstein

, Gambatese

, Hecker

. Can design improve construction safety? Assessing the Impact of Collaborative Safety-in-Design Process, Journal of Construction Engineering and Management. 2005;131:1125–34.

35.

Workplace Safety and Health Council –WSH Council [homepage on the Internet]. Singapore; 2008. [updated 2011 October 19 cited 2011 October]. Available from: https://www.wshc.sg/wps/portal/accidentCaseStudies.

36.

WSH Council. [homepage on the Internet]. Singapore; 2008. Guidelines on design for safety in buildings and structures. [updated 2009 April 19 cited 2009 April]. Available from: https://www.wshc.sg/wps/themes/html/upload/cms/file/Guidelines%20on%20Design%20for%20Safety%20in%20Buildings%20and%20Structures.pdf.

37.

WSH Council. [homepage on the Internet]. Singapore; 2008. Healthy Workforce. Safe Workplace. [updated 2009 April 19 cited 2009 April]. Available from: http://www.wshc.sg/wps/portal/accidentCaseStudies.

38.

Zhu

, Hwang

, Ngo

, Tan

JPS

. Applications of smart technologies in construction project management, Journal of Construction Engineering and Management. 2010;148(4):040220101–04022010-12.