How to measure safety in construction industry?

Too many accidents happen in the construction industry globally. Retrospective methods are used to understand the causes and to reduce the number of accidents. However, the results are limited by imperfections in accident recording, and in research and analysis of accident causes. Therefore there is need for a prospective method to detect potential causes of accidents and to reduce the risks by the introduction of safety interventions. When (un)safe conditions of apparently normal situations are quantified, improvements can be made before accidents actually happen. Five methods to identify unsafe conditions are compared, regarding their usefulness in the construction industry; the ‘TR safety observation method on building construction’ (Finland), the `Injury Exposure Assessment’ (USA), the `benchmark method’ (Australia), the `checklist safety indicator’ (Australia) and the `Disturbance Assessment method’ (The Netherlands). The `TR safety observation method on building construction’ is considered to be the most suitable method. The method is validated and easy to use in the operational phase of the building process. It is based upon a scenario approach. Comparing various workplaces is easy with this method and the results will inspire the introduction of safety interventions. This paper describes the results of a comparative research into methods and a limited field test, using the Dutch version of the TR Method. A more extended field test is being carried out to assess the suitability of the method for the entire construction industry in the Netherlands.
Palabras Clave: 
building
Autor principal: 
Adri C.P.
Frijters
Coautores: 
Yperen
Hester R. van
Paul H.J.J.
Swuste

Arbouw / Postbus 8114 / 1005 AC Amsterdam / Frijters@arbouw.nl

Paul H.J.J. Swuste

Research group on Safety / TU Delft

Hester R. van Yperen

Lumnos bv, scientific communication on toxicology and health matters

ABSTRACT

ABSTRACT

Too many accidents happen in the construction industry globally. Retrospective methods are used to understand the causes and to reduce the number of accidents. However, the results are limited by imperfections in accident recording, and in research and analysis of accident causes. Therefore there is need for a prospective method to detect potential causes of accidents and to reduce the risks by the introduction of safety interventions. When (un)safe conditions of apparently normal situations are quantified, improvements can be made before accidents actually happen.

Five methods to identify unsafe conditions are compared, regarding their usefulness in the construction industry; the ‘TR safety observation method on building construction’ (Finland), the `Injury Exposure Assessment’ (USA), the`benchmark method’ (Australia), the `checklist safety indicator’ (Australia) and the`Disturbance Assessment method’ (The Netherlands). The `TR safety observation method on building construction’ is considered to be the most suitable method. The method is validated and easy to use in the operational phase of the building process. It is based upon a scenario approach. Comparing various workplaces is easy with this method and the results will inspire the introduction of safety interventions.

This paper describes the results of a comparative research into methods and a limited field test, using the Dutch version of the TR Method. A more extended field test is being carried out to assess the suitability of the method for the entire construction industry in the Netherlands.

Keywords

Keywords

Construction industry, TR Method, measure safety

INTRODUCTION

INTRODUCTION

Working in the construction industry means being involved with safety issues. The Dutch Economic Institute for the Building Industry (EIB) has estimated that there were 9250 work-related accidents in 2006 (Blomsma, 2007). Furthermore, both employers and employees have noticed over a period of many years that there are too many accidents in the building industry.

In May 2004 the construction industry began the campaign ’1 in 6’ to make builders more conscious of the risk of falling. The slogan 1 in 6 refers to the estimated risk of falling during a career in the building industry. This campaign has helped to make clear that there is a need for valid and user-friendly measuring methods so that everyone becomes aware of the dangers on a building site and takes the necessary precautions.

This paper describes research into methods of quantifying safety levels on building sites. To answer this question it was necessary to discover which methods are available for doing this, which method would be most suitable and whether it would be applicable on Dutch building sites. These methods originate from various countries and most of them have been tested for user-friendliness, for extent of use and for reliability. A detailed justification for carrying out this research has already been published by Arbouw (Frijters et al. 2007).

METHODS AND TECHNIQUES

Surprisingly few methods for determining safety levels on building sites have been described in the scientific and professional literature. Five different methods were selected for this research (Folkerts, 2006). One of the methods comes from Finland (the TR method), one from the US (the Injury Exposure Assessment), two from Australia (the Benchmark method and the Checklist of Safety Indicators), and there is one from the Netherlands (the Disturbance Analysis Method).

TR Method

This method, officially called the ‘TR safety observation method on building construction’, was developed by the Finnish Institute of Occupational Health as an audit method for building site workers. (Laitinen and Ruohomäki, 1996; European Agency for Safety and Health at Work 2004). The abbreviation ‘TR’ is the Finnish acronym for ‘building site’. The method was tested on 200 building sites, validated, and found to be good enough to be widely implemented in the building industry. The Labour Inspectorate in Finland is working with the method. It seems a high score for the audit -- in other words, a high level of safety -- corresponds with low accident totals at the building sites involved (Laitinen et al, 1999).

The TR method is an observation and assessment method which allows safety on building sites to be ‘measured’ by building site personnel. The measuring method draws attention to both effective and ineffective control measures and, to a lesser extent, to both desirable and undesirable behaviour (table 1).

Name of company:

Building site:

Address of building site

Phase of work: foundations, structural work, completion

Component

Score ’correct’

Total

Score ’incorrect’

Total

1. work practices

2. scaffolding and ladders

3. machines and tools

4. fall protection

5. light and electricity

6. workplace

Total number ‘correct’

Total number ‘incorrect’

Safety index = total correct / total number observations X 100% =

Name observer

Signature

Date


Comments


The observer looks at each component of the work practice to see if the criteria for safety or safe usage are being met. An explanation of these criteria is shown in detail in table 2. Situations are given the evaluation “correct“ if they meet the safety criteria and “incorrect“ if they fail to do so.

Component

Number of observations:

Criteria used for giving a ‘correct’ score

Work practices

Use of personal protection measures

Taking risks while carrying out work

- 1 per employee

Proper use of necessary PPE (Personal Protective Equipment)

No really obvious risks taken

Scaffolding and ladders Scaffolding, ladders, ladder frame towers, trestles

- 1 for every scaffold construction, stand-alone ladder or temporary working place high off the ground

Equipment stands firmly

Sufficient anchorage (safety equipment, anchors)

Floor has no openings

Construction is adequate for the job Constructed according to manual

Space between temporary floor and walls kept to minimum

Machines and tools

All types of machines and tools in use

- 1 for every type

Condition of machine/tools (evidence of inspection, visual inspection shows machine/tool in order)

Working environment does not increasehazards

Set up firmly and safely

Normal work postures can be maintained

Fall protection Floor edges Floor boxes

Earth removal sites Excavations/pits Permanent stairs Holes

1 for each vertical distance that requires safety measures

Edges protected with some form of ‘edge protection system’: handrails, midrails, toeboards

Floor boxes closed up or provided withedge protection systems

Earth removal sites protected by bank wall or other means of preventing collapse Excavations and pits given edge protection Stairways and stairwells protected Measures to prevent protective covers slipping or being removed

Light and electricity

Total lighting

Lighting on walkways and entry roads

Lighting on stairs and ladders

Lighting of workplace

Use of cables and temporary power sockets (distribution boards)

1 for total lighting 1 for total light on walkways and entryroads, ladders and stairs1 for each workplace in use1 per monitoring point forcables and temporary power sockets

Artificial light is acceptable provided the point of work is evenly lit

No shadows should be cast on walkways,floors and stairways (there is increased risk)

Cables and power points should not be in water

No possibility of touching anything that transmits electricity (risk of contact)

workplace

How workplace is organized Order and tidiness in workplace

Order and tidiness onwalkways and entry roads Danger of tripping; accessibility

Waste bins

Traffic diversion

1 for each workplace 1 per monitoring point1 for order and tidiness, and use of waste bins 1 for marking off work area1 for information signs in& marking out of work area

Workplace well organized No danger of tripping

No unnecessary building waste Rubbish / waste separated

Transport, paths and roads immediately usable

Marking off and signing roads according to CROW guidelines

The number of assessments on the form is not the same for each component. The scenario “falling from a height“ is given a score under the heading “scaffolding and ladders“ but also under the heading “fall protection “. In other words, each type of ladder and each vertical distance is evaluated separately.

Consequently, the scores for danger of falling have a higher weighting in the final score than the other components. This emphasis on the danger of falling can be justified by the high incidence of accidents related to falling. (Dutch Labour Inspectorate, 2007, Blomsma, 2007).

The observer makes at least 100 observations at each building site. All these observations lead to one score for safety, the safety index. This is the relationship between the number of observations that have been scored “correct“ and the total number of observations. The index can vary from 0 to 100% and is easy to interpret. It is also easy to see which work component has contributed to which part of the total score, thus allowing problems to be targeted and solved effectively.

Injury Exposure Assessment

The Injury Exposure Assessment (IEA) method allows an estimation to be made of the risk of an accident occurring during the building implementation phase (Seixas et al., 1998). An observer visits a building site and chooses a random number of observation points, anything between 10 and 40. He then scores the various dangers using a checklist (see table 3), taking into account the presence and degree of protection.

Table 3: Elements of the risk checklist used in the IEA method (Seixas et al. 1998).

Item

Type of injury

Definition of risk

1

Trips

Rough, irrigular walking surface or materials/ tools/ debris likely to result in tripping

2

Minor fall

Vertical work (e.g. stepladder) or work surface edge with potential for a fall of less than 6 ft

3

Major fall

Vertical work (e.g. ladder) or work surface edge with potential for a fall of more than 6 ft

4

Electrical distribution

Presence of any electrically charged materials, including distribution circuits, electrical cords, wires, orpower lines (excludes electric-powered tools or equipment and totally enclosed power lines such as within

a wall)

5

Electrical powered tools

Electric tools or equipment such as lights

6

Trenching/ excavations

Trench or earth excavations large enough to allow human entry and at least 4 ft deep

7

Mechanical impact

Potential for impact from machines, tools, or oving objects, including objects falling from heights

(excludes vehicles)

8

Vehicles

Potential for contact with a vehicle, including both operator and workers in the vicinity of operation;

vehicle must be present

9

Thoroughfare

Roadway or passage normally used by construction vehicles around, through, or within site

10

Cuts/ impail

Presence of an object that could cut or impal if worker is in contact

The degree of protection is estimated using a weighting between 0 and 10. A0 means no protection while 10 stands for maximum protection. The results are then subsequently added together for each danger. By means of this method, the observer can evaluate whether there may be a risk situation. The IEA has been tested on three building sites, but not validated. Tests showed that there were large and significant differences in the assessments made by different observers. This seems to partly depend on the level of expertise of the observer, so testing randomly chosen observations has its drawbacks. High risk situations which do not happen very often will probably remain undiscovered using this method.

Benchmark Method

The Benchmark Method introduces the ‘Balanced Scorecard (BSC)’, a strategic management tool (Mohamed, 2003), having the following 'measuringpoints':• the customer perspective• potential for innovation or learning something new• company organization/strategy perspective• company results perspective

The BSC provides a method of screening a company for a wide range of features. The way in which a company functions can then be compared with the situation in other companies. The investigation of this method led to a series of aims and performance measurements for each of the four measuring points (table 4).

Table 4: Suggested goals and performance measures used in the Benchmark Method

Perspective

Suggested Goals

Basis of suggested performance measures

Qualitative

Quantitative

Management

Eliminate accidents Reduce incidents Improve productivity

Number of accidents Number of incidents

Degree of performance reliability

Lead by example (management commitment) Reduce accident-related cost

Extent of management involvement to improve safety Dollars saved on accidents reduced

Emphasize subcontractors' safety-awareness

Number of safety issues "pushed" down to subcontractors

Learning

Continue to improve safety performance level Build highly competent workforce

Number of safety initiatives

Extent of ability to transfer learning into workplace

Empower workforce

Establish an effective strategic feedback system

Extent of workforce proactive involvement to improve safety Number of safety audits/ reviews

Provide adequate training to new recruits

Number of hours of competency/ induction training

Operational

Establish and maintain a safe workplace Establish an operational feedback system Implement an efficient follow-up system Carry out more effective site layout planning Create a better working environment

Score of compliance/ noncompliance to safety requirements Score and/ or number of safety audits/ focus groups/ reviews Recommended/ implemented remedial actions ratio

Number of incidents due to poor safety integration into planning

Degree of satisfaction with currect working relationships,

safe behaviours and attitude towards safety

Customer

Ensure client satisfaction Instigate employee satisfaction

Exceed project partners expectations

Client satisfaction rating

Number of complaints/ grievances/ legal suits Extent of meeting/ exceeding their expectations

Enhance workforce morale

Extent of recognizing and rewarding individuals with

excellent safe performance

The Benchmark Method closely resembles a professional audit; it supplies specific and usable data about the safety status of the building process. There are a number of suggestions in the literature for using the method at an operational level, but the method is still at the research stage and has not yet been fully tested and validated.

Checklist Safety Indicators

Many people believe that accident statistics do not have much in common with safety intervention (Hale, 2005). These statistics are usually published in Annual Reports but few construction companies make use of the information in order to start up safety campaigns (Duff, 2000; Laitinen et al., 1999). This observation led to the Checklist of Safety Indicators; a method for making safety quantifiable and for using the results to implement improvements so that the safety levels of a company can be increased (Trethewy, 2003).

In total, 58 indicators have been developed for various phases of a building project. The indicators are scored on a six-point scale varying from zero (0) to excellent (5).

These 58 indicators are divided over the phases from design to implementation as follows:

  • 1. drafting and feasibility phase (2 indicators),
  • 2. design and planning phase (10 indicators),
  • 3. selection and invitationtotender phase (4 indicators),
  • 4. implementation phase divided into 4 subphases (38 indicators),
  • 5. completion, maintenance (4 indicators).

The Checklist has the characteristics of an audit. Measurement covers all aspects and is done thoroughly, the whole building process and individual companies being scrutinized. The Checklist, which provides comprehensive and specific data about the building process, has not yet been tested, so it is not possible to comment on its reliability.

Disturbance Analysis Method

The Disturbance Analysis Method is based on HAZard and OPerability study (HAZOP) and is aimed at tracing all possible undesirable events or defects in future processes. A group of experts from different backgrounds take part in structured brainstorming sessions. With the help of drawings and presentations, they carefully go through the whole process looking for possible process disturbances. In order to do this, they use quantifying terms (‘guide words’) such as ‘more’, ‘less’, etc. These guide words represent deviations from the norm and can be coupled to process parameters such as pressure, temperature and materials. These parameters represent dangerous aspects and can be combined in a matrix as shown in table 5.

Table 5: Example of a Disturbance analysis matrix

PROCESS PARAMETERS

Applicable guide words

None

More

Less

Direction

Reverse

Other

Activity/Pressure

Space

Movement

Time

Speed

Energy

Materials

Relevant combinations are looked for; in other words, there is a realistic possibility of deviation according to the experts, or the data have already been registered during incidents and disturbances. After that, a scenario is evaluated as occurring 'with a high frequency' or 'with a low frequency'. The matrix then results in a series of scenarios.

The results of testing this method in the steel industry were encouraging (Swuste et al, 1997). The scenarios that had been predicted really did take place. However, there was less success when testing was done on road building activities (Swuste et al, 2000). This may partly have been due to a limited period of working in the field and to the quality of the presentations used.

Comparison of the five measuring methods

The measuring methods were compared with each other using the aspects listed below:

• In which part of the building process can the method be applied?• Does the method make use of various scenarios?• What information is presented in the results?• How is safety measured (scale of measurement)?• Who can implement the method and how easy is it to use?• How long does the method take to implement and how often is this done?• Has the method been validated?

After evaluating the five methods, one of them was further developed for use on Dutch building sites.

This method was then tested for ease of understanding and implementationin the Netherlands by getting it to be used on various building sites.

RESULTS

Comparison of the five measuring methods

Table 6 shows a comparison of the five methods. Table 6: Comparison of measuring methods


TR Method

Injury Exposure

Benchmark method

Checklist Safety

Disturba nce

Assessment

Indicators

Analysis

1

Which part of the

During

During

Whole

Whole

New

building process?

operational

operational

building

building

process

level

level

process

process

being

started

2

Based on

Yes

Yes

No

No

Yes

scenarios?

3

Results after

Relative

Relative

Comprehen

Relative

Possible

implementation?

score after

score after

sive scan of

score after

scenario

evaluating

evaluating

building

evaluating

s when

scenarios

scenarios

company/p

58

impleme

rocess

indicators

nting

on

process

processes

and safety

4

Measuring scale?

correct/

10 points

n.a.

5 points

n.a.

incorrect

5

Level of method?

Simple

Simple

Complicate

Complicate

Complic

d

d

ated

6

Who carries out

SHE

SHE

SHE

SHE

Team of

measurements?

personnel

personnel

personnel /

personnel /

experts

and building

manageme nt*

manageme nt*

site

personnel

7

Much time

½ - 1 hour

1 hour*

Half a day*

Half a day*

Half a

needed for

day

measuring?

8

Has the method

Yes

Only tested

No

No

Only

been validated?

tested

*assumption

The comparison showed that the TR method was the most suitable for use on Dutch building sites. This was the only method to be validated and was also the simplest to implement for building site personnel. It was therefore decided to make the TR method suitable for the Dutch situation. This resulted in the Safety Indicator which was then further tested on a number of building sites.

Testing the Dutch Safety Indicator

The TR method was translated to produce a Safety Indicator which was subsequently tested on four types of building site: new developments, reconstruction, working with prefab units and renovation (table 7). On one of the building sites, the method was applied twice, with a period of time in between. It was decided to choose a building site where the employees from one single construction company plus sub-contractors worked. This choice was made for logistical reasons; the chosen construction company had sufficient large building sites located near to Eindhoven. The annual turnover of this construction company is around €250 million. The monitoring team consisted of building site personnel, management and KAM consultants.

Table 7: Overview locations where tests were done

Test number

Building project

Test

1

New development/renovation senior citizens’ centre

Test

2

New development 125 rented apartments (high rise)

Test

3

New development/new land-use plan involving 37 owner-occupied apartments and basement parking

and

rented

Test

4

New development 109 owner-occupied apartments (high rise)

Test

5

New development Music Centre

Test

6

New development/renovation senior citizens’ centre

The Safety Indicator was introduced to the company during a meeting for all building site employees. Method and evaluation were explained using a number of examples. During the field trials, the researcher had fulfilled the role of helpdesk whenever things were not clear. After the completion of the test, a quick evaluation was held. The results of the field trials are shown in table 8.

Table 8: Results of field tests

component

Test 1

Test 2

Test 3

Test 4

Test 5

Test 6

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

Correct

Incorrect

Work practices

9 6

7 0

7 2

41 2

14 15

12 3

Scaffolding and ladders

24 6

8 3

9 6

4 0

17 11

7 2

Machines and tools

9 0

1 0

6 2

5 0

11 2

10 1

Fall protection

19 7

16 15

18 7

69 5

8 11

12 19

Lighting and electricity

5 1

2 1

14 2

23 0

8 0

11 3

Order and tidiness

11 4

5 4

28 4

85 1

11 16

21 2

Time taken (mins.)

55

50

45

105

55

45

Number ‘correct’

77

39

82

227

69

73

Number of observations

101

62

105

235

124

103

Safety index

76%

63%

78%

97%

56%

71%

Experiences in implementing the method

Only a brief explanation seemed to be necessary for all participants to understand the method. However, one of the tests consisted of fewer than the required 100 observations and in some cases a dangerous situation (lack of fall protection, untidy working area) was not recognized as unsafe by the person doing the monitoring. And during one of the tests – on a high rise building project – each floor and apartment was included in the monitoring process. This resulted in many observations, but it also took a long time to carry out the method (105 instead of 60 minutes).

Although the instructions stated: "if in doubt, do not score", some observers still gave “correct“ or “incorrect“ evaluations even if there was some doubt. One of the observers even mixed up the various observation components. Also, multiple observations were sometimes included in a single score. In addition, some observers asked the researchers if it was possible to give a "nearly correct" score because it was sometimes difficult to give a total disqualification.

It seems the observation forms were not completely satisfactory. The data fields were too small and it was awkward working with more than one sheet of A4. Furthermore, ordinary paper was not suitable for the working and weather conditions on a building site.

Experiences with the six observation components

Work practices

Most of the observers seemed to be able to assess whether somebody was working safely or not. However, one of the observers failed to notice that some people were working on high constructions without fall protection.

Scaffolding and ladders

The observers pointed out that in some cases it was unclear whether a situation was safe or not.

Machines and tools

The observers seemed to find it difficult to observe this component. It was particularly difficult to ascertain if the equipment had been approved or not. Some observers did not know enough about this subject. Mostly, they just looked at whether the machines were being used correctly or not.

Fall protection

Various observers indicated that they wanted to assess some situations as “almost correct”. Another problem was that not every observer had sufficient knowledge of the regulations regarding fall protection, so from time to time, dangerous situations such as open floor boxes, were not noticed.

Lighting and electricity

In situations where cables had been laid in a disorderly fashion, the assessments of whether this was correct or incorrect were inconsistent. On the other hand, assessments of lighting installations were done consistently.

Order and tidiness in the workplace

Most observers tended to base their evaluations on their own personal opinions.

CONCLUSIONS AND DISCUSSION

Of the five safety measurement methods investigated, the TR method appeared to be the most suitable one. It is the only validated method and it can be used by building site workers themselves. In addition, it takes up relatively little time and results in data in the form of a Safety Index that is easy to interpret. The results can easily be communicated with building site employees and management during normal meetings. This feedback is expected to increase commitment to safety and to reduce the number of unsafe situations on building sites.

The Safety Indicator is used to measure the quality of the barriers -- the safety measures. If these barriers receive a positive score, then accidents are prevented or reduced by the barriers. This effect should be visible in the accident registrations of the companies concerned and so, in this way, the effectiveness of intervention is more noticeable.

The TR method has much in common with the IEA method but the complexity of the latter makes it unsuitable for use by building site employees. A second difference is the type of measuring scale which in the case of the IEA method is a 10-point scale, whereas the TR method has a 2-point scale. The latter, where the score is either correct or incorrect, is not sensitive enough to show small changes (Duff, 2000), and there is a disadvantage in that the use has only limited possibilities for evaluating hazards. This showed up in the tests and indicated that a more detailed explanation is needed when the method is introduced. On the other hand, there is more chance of subjectivity with a multipoint scale score than with a 2-point scale. During the investigation, the Dutch version of the TR method was implemented at a limited number of building sites belonging to just one company. The emphasis was on the usability of the method, so it did not result in a systematic overview of the validity and reliability of the method. It will certainly be necessary to further investigate the possible differences in scores between different observers, as well as to compare the results from the TR method with those from other safety methods.

Although the TR method was tested on only limited numbers and types of building site, it could also be suitable for use in other sectors of the industry, provided the method is explained properly and a few adjustments are made.

The TR method is not meant to be a replacement for existing safety management systems. However, it does allow the safety level to be made more visible. It can also form part of the decision-making process regarding safety measures.

At the request of the sector, Arbouw is currently carrying out a pilot test using the Safety Indicator at various types of building site. The test involves some150 Dutch building sites belonging to around 75 companies, and it will result in more data about usability and about the best way of making the Indicator available. The results of this pilot test should also be useful for comparing the safety levels of different building companies.

LITERATURE

  • 1. Arbeidsinspectie, (2007) jaarverslag 2006, Den Haag, Ministerie van Sociale zaken.
  • 2. Blomsma  G.,  Lourens  E.  (2007)  Arbeidsongevallen  in  de  bouw  in  2006 Arbouw Amsterdam
  • 3. Deming W.E. (1991) Out of the Crises, Cambridge, MIT Press
  • 4. Duff A.R. (2000) Behaviour Measurement for continuous improvement in construction safety and quality, in: Richard J. Coble (eds), The management of construction safety and health Balkema, Rotterdam
  • 5. European Agency for safety and health at work (2004) Achieving better safety and health in construction. ISBN 9291910732, Office for official publications of the European Communities, 2004, 6572
  • 6. Hale A., (2005) Safety Management, what do we know, what do we believe we    know and what do we overlook? Tijdschrift  voor  Toegepaste Arbowetenschap,  18(3);  5866
  • 7. Folkerts H. (2006) Veiligheid meten op de bouwplaats. Een studie naar de mogelijkheid van gebruik van een Finse Veiligheidsmeetmethode voor Nederlandse bouwplaatsen. Verslag van een eindstudie in de richting Uitvoeringstechniek. Technische Universiteit Eindhoven, faculteit Bouwkunde
  • 8. Frijters A.C.P. Swuste P.H.J.J. Yperen H.R. van (2007)  Het  meten  van veiligheid op de bouwplaats, de veiligheidsindicator (Arbouw, Amsterdam
  • 9. Kroonenberg H.H. van den, en F.J.Siers (1992) Methodisch ontwerpen: ontwerpmethoden, voorbeelden, cases en oefeningen Educaboek, Culemborg
  • 10. Laitinen H. Ruohomäki I. The effects of feedback and goal setting on safety performance at two construction sites, Safety Science 1996:24(1);6173
  • 11. Laitinen H. Marjamäki M. Päivärinta K. The validity of the TR safety observation method on building construction. Accident Analysis and Prevention   1999:31;463472.
  • 12. Lund J. Aarø L.Accident  prevention.  Presentation  of  a  model  placing emphasis on human, structural and cultural factors, Safety Science 2004:42;271324
  • 13. Mohamed S. Scorecard Approach to  Benchmarking  Organizational  Safety Culture  in  Construction,  Journal  of  Construction  Engineering   and Management   2003:129(1);8088
  • 14. Saari J. The Effect of positive feedback on industrial  housekeeping  and accidents;  a  longterm  study  at  a  shipyard,   International.   Journal   of Industrial  Ergonomics  1994:4;201211
  • 15. Seixas N.S. Sanders J. Sheppard L. Yost M. Exposure assessment for acute injuries  on  construction  sites:  conceptual   development   and   pilot   test, Applied Occupational Environment Hygiene 1998:13(5);
  • 16. Swuste P. Goossens L. Bakker F. Schrover J. Evaluation of accident scenarios in a Dutch steel works using a hazard and operability study. Safety Science   1997;26(12):6374
  • 17. Swuste P. Wiersma E. Frijters A. Storingsanalyse voor Arbeidshygiënisten?, Tijdschrift voor toegepaste Arbowetenschap, 2000:13(2); 1823
  • 18. Trethewy R.W. OHS Performance Improved Indicators for Construction Contractors, Journal of Construction Research 2003:4(1);1727

Papers relacionados

ORP 2015
Adolfo Guillermo
Cortes Rivera
ACHS - Asociación Chilena de Seguridad
Chile
ORP 2015
Francisco José
Forteza Oliver
Universidad Illes Balears
España
ORP 2015
Isidro J
Ibarra Berrocal
Universidad Politécnica de Cartagena
España
ORP 2015
Iván de Jesús
Arboleda
Alcaldía de Medellín
Colombia
ORP 2015
Macringer
Omaña
Universidad Simón Bolívar
Venezuela
ORP 2015
JEILENE ABIGAIL
ROSAS SALAS
INSTITUTO POLITÉCNICO NACIONAL
México