Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 172 (2017) 355 – 362

Modern Building Materials, Structures and Techniques, MBMST 2016

Methodology of Analysing the Accident Rate in the Construction

Industry
Bożena Hołaa, Mariusz Szóstaka *
a
Wroclaw University of Technology, Faculty of Civil Engineering, Department of Construction Methods and Managements,
Wybrzeże Wyspiańskiego 27, building C-7, Wrocław 50-370, Poland

Abstract

The article presents the methodology of analysing and evaluating the accident rate in the construction industry, which consists
of five stages: the identification of sources of data on accidents at work and the acquisition of research material, the classification
of research material, the formulation of a computer knowledge base, the creation of a model which simulates the course of the
complex process of an accident and also the execution of calculations and analysis of obtained results. An example of analysis
of an accident process with 130 accidents at work was also provided.
© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-reviewunder
Peer-review under responsibility
responsibility of organizing
of the the organizing committee
committee of MBMST
of MBMST 2016 2016.

Keywords: construction industry, accident at work, accident rate, scenarios of accident rate development.

1. Introduction

Occupational safety in specific sectors of the economy is mainly evidenced by the number of severe and fatal
accidents. In Poland in 2014, 88,000 accidents at work occurred in all sectors of the economy, in which 263 people
were killed. The construction industry has a very high place in the ranking of individual sectors with regards to the
accident rate. In 2014 this ratio was equal to 7.68 injured people per 1,000 workers. This value is higher than the
accident rate obtained for all sectors of the national economy, which was equal to 7.45 injured people per 1,000 workers

* Corresponding author. Tel.: +48 71 3202369; fax: +48 71 3221465.

E-mail address: mariusz.szostak@pwr.edu.pl

1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of MBMST 2016
doi:10.1016/j.proeng.2017.02.040
356 Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362

[1]. Therefore, the subject of the accident rate in the Polish construction industry should be considered as important
and there should be a lot of attention devoted to it.

2. Description of the research problem

Information regarding accidents at work is published by state institutions. This is mostly statistical data on accidents
in various sectors of the economy. From these publications we can find out e.g. how many people were injured
in accidents at work in the construction industry, how many people were injured at a construction site and other places,
how many people at the time of the accident were using a machine and also how many people were electrocuted.
On the basis of this data, however, any conclusions about the most probable course of accidents and situations
on a construction site, which pose the greatest risk to workers, cannot be drawn. This is because the relations that exist
between specific data sets are not known [2].
Detailed information about the course of an accident can be found in the accident-investigation protocol drawn
up by the labour inspector after its occurrence. Based on this information, the course of a single accident at work can
be reconstructed. However, knowledge of a single accident is not enough to draw conclusions about the sustainable
features of the phenomenon of the accident rate. Such conclusions can be drawn on the basis of results obtained from
research on an appropriately numerous set of accidents.
The objective of the carried out study was to develop a methodology of the analysis of the accident rate in the
Polish construction industry, which enables the most probable mechanisms of formation and development of accident
situations to be defined. Based on preliminary research and analysis of accident protocols the following assumptions
were made:

x The accident rate is a mass phenomenon and the state of safety in specific sectors of the economy is not determined
by a single accident but by the set of accidents that occurred within a specified period of time, and also information
which resulted from analysis of this set.
x Accidents at work happen at certain moments of time and create a chain of events that can be analysed as a complex
discreet process.
x Each accident occurs according to a specific scenario.
x Appropriately directed analysis of a process that consists of many accidents will enable to be defined the space
of possible scenarios of accident situation development, the probability of occurrence of certain scenarios and also
scenarios with the highest probability of occurrence.

3. Methodology of research

Original methodology of the analysis of the development trend of the accident rate in the construction industry was
developed and consists of the following stages:

x Stage I – the identification of sources of data on accidents at work and the acquisition of source documents.
x Stage II – the selection of a representative set of documents and their analysis. Classification of the obtained data
and information.
x Stage III – formulation of a knowledge base on accidents at work.
x Stage IV – creation of a model of the development trend of the accident rate which is based on graph theory.
x Stage V – studies on the model and analysis of obtained results.

The general scheme of the proposed methodology of analysis is shown in Figure 1.

Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362 357

Fig. 1. General scheme of the developed methodology.

358 Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362

3.1. Stage I

Information about the course of accidents at work can be found in the archives of the State Labour Inspectorate
(SLI), the Central Statistical Office (CSO), the Social Insurance Institution (SII), the prosecutor's office, courts and
also insurance companies. However, access to this data is very limited because:

x They contain confidential personal data of injured people.

x In most institutions all sets of data are very large and are organized in a way that does not allow the rapid
identification of accidents that have occurred in the construction industry.
x A search through such sets is time consuming and involves considerable financial outlay.

Despite these difficulties, documents in the form of accident-investigation protocols were possessed from the
collections of the State Labour Inspectorate. In order to protect the personal data of people included in documents, the
documents have already been rendered anonymous.

3.2. Stage II

The study involved accidents that happened in the Polish construction industry in the time period between 2008
and 2014. Over 1,200 control protocols on accidents at work in the construction industry were received from individual
archives of the State Labour Inspectorate. On the basis of a selected representative sample, which amounted to 130
protocols, a classification of the content included in them was made. The structure of knowledge about accidents at
work was created and included:

x general information about an accident such as: date, place and the size of an enterprise where the accident occurred,
x details of the injured person: age, form of employment, occupation, experience,
x information about the course of an accident: the place of an accident, the type of work carried out by the victim,
the operation performed by the victim at the time of an accident, the material factor associated with an action
performed by the victim at the time of an accident, an event incompatible with the appropriate course of work, the
material factor associated with a deviation, an event causing injury, the material factor which is the source of injury,
the type of injury, the result of an accident, causes of an accident divided into technical, organizational and human
causes.

3.3. Stage III

The structure of knowledge, which was developed in Stage II, was the basis for the construction of a computer
knowledge base of accidents. This database is built in the form of a two-dimensional table. Each row contains
information about one accident. Designations included in the dictionary of terms used in the publications of the Central
Statistical Office were used to describe accidents. Each defined designation also has a numeric code assigned to it.
The matrix form and used numeric codes allow the database regarding interesting information to be easily searched
through. The knowledge base is a repository for collected data and enables various statistical analyses of a set of
accidents contained in the database to be conducted.

3.4. Stage IV

Each case can be considered as a process consisting of a pre-accident phase, the phase of the accident and post-
accident phase [3]. The pre-accident phase includes elements of the work process, which relates to an employee and
the tasks performed by him at the time of an accident. The accident phase is separated from the pre-accident phase by
an event that is a deviation from the normal state and is inconsistent to the appropriate course of the work process
which caused the accident. The deviation is generally related to a material factor, which can be a machine, tool
or another object, or to an environmental factor that has a direct relation with an event that is a deviation from the
 Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362 359

normal state. The source of an injury is the material factor, which became the direct cause of the injury. The post-
accident phase deals with the consequences of the accident.

Fig. 2. Model of an accident at work.

Based on analysis of accident-investigation protocols, it was concluded that each accident occurs according
to a different scenario. The set of accidents can be presented in the form of a directed graph (Figure 3). Nodes in the
graph represent the elements of the work process which were extracted from the accident process, and the arcs
represent the relationships between different elements. Each accident corresponds to one path in the graph.

Fig. 3. Model of accident rate development in the construction industry.

360 Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362

The computer model of the development of an accident situation in the construction industry, which is based on the
principles of graph theory, is created in stage IV [4]. This model allows the course of each accident, which is a part
of the analyzed complex process, to be simulated.

3.5. Stage V

Calculations on the constructed model are carried out in stage V. A sequence of accident events, that includes
accidents that occurred within a defined period of time, is implemented into the model. Courses of successive accidents
are simulated in the model according to the individual accident scenarios stored in the computer knowledge base.
Characteristics of the accident process in the form of the frequency of initiating individual arcs, the probability
of a certain scenario occurrence and the identification of the critical path, therefore the process with the highest
probability of initiation, are obtained at the output of the model.

4. The computational example

A simulation of an accident process which consists of 130 accidents at work that occurred in the construction
industry in Lower Silesia between 2008 and 2012 [5] was conducted. Subsequent accidents, which form the analyzed
process (Fig. 2) and are downloaded from the computer knowledge base, run according to the individual path that
starts at the node "START" and ends at the node „ws” (Fig. 3). Designations of nodes in the graph relate to the structure
of information contained in the knowledge base:

x place of an accident – a building site: a021 - a building object under construction; a022 - a building object under
renovation; a024 - underground; a025 - on water, above water; a026 - in a high pressure environment.
x type of work performed by a victim – mining and excavation works, construction works: b21 - mining and
earthworks; b22 - building new buildings; b23 - implementation of infrastructure, e.g. roads, bridges, dams, ports,
etc.; b24 - rebuilding, repair, development, maintenance of building objects and infrastructure; b25 - dismantling and
demolition works.
x an operation performed by a victim at the time of an accident: c1 - using machines; c2 - work with hand tools;
c3 - steering/driving means of transport/servicing moving machines and other devices; c4 - handling of items;
c5 - manual transport; c6 – movement; c7 - presence.
x events which are a deviation from the normal state: d1 - deviations associated with electricity, an explosion, fire;
d2 - discharge, spillage, emission of harmful substances; d3 - damage, bursts, cracks, slipping, falls, failure of the
material factor; d4 - loss of control over a machine, means of transport, transported cargo, tools or objects;
d5 - slipping, stumbling, falling off; d6 - movements of the body without physical effort; d7 - movements of the
body associated with physical effort; d8 - shock, fear, violence, attack, threat, presence.
x the event causing an injury: e1 - contact with electric current, temperature, hazardous substances and chemical
solutions; e2 - drowning, burial, closure; e3 - collision with an immovable object; e4 - being hit by an object
in motion; e5 - contact with a sharp, rough, or stippled object; e6 – being trapped, crushed; e7 - physical or mental
stress; e8 - human aggression.
x the type of injury: f010 - wounds and superficial wounds; f020 - broken bones; f030 - displacements, dislocations,
sprains and tears; f040 - amputations (losses of body parts); f050 - internal injuries; f060 - burns with fire or chemicals,
burns with water or steam, frostbites; f070 - poisoning, infections; f080 - drowning, suffocation due to lack of oxygen;
f090 - effects of sounds, vibrations and pressure; f100 - the effects of extreme temperature, light and radiation;
f110 – shocks; f120 - numerous injuries; f130 - injury caused by a fall from a height; f140 - injury as a result of backfilling
with soil; f150 - the death of a victim.
x the result of an injury: w1 - death of the injured person; w2 - serious body injuries; w3 - light body injuries.

Based on the results of the simulation of the process which was composed of 130 accidents that were classified
as fatal and serious, an area of possible scenarios of accident events was defined and is shown in Figure 4. The course
of the critical path is marked in red. From this Figure the following can be concluded:
 Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362 361

x Accidents at work occurred most frequently during the construction of new building objects (a021) - buildings (b22)
- and also during the renovation of existing building objects (a022). There were no accidents during construction
works underground, e.g. in mines or during the construction of tunnels, and also on water or above water, e.g.
during the construction or renovations of oil rigs at sea.
x An accident occurred most frequently during the movement of an employee (c6), i.e. walking along a flat terrain,
ascending or descending to another level, entering or leaving other spaces.
x The events that were incompatible with the proper course of work were slipping, tripping or the falling
off of a person onto a lower level or falling over at the same level (d5), which was followed by a vertical collision
with a fixed object or the hitting of an immovable object (e3).
x As a result of these events, the most common type of injury were bone fractures (f020) leading to the severe injury
of a victim's body (w2) or the death of a victim (f150), (w1).

The results of a preliminary analysis of accidents at work in Poland do not differ from the results obtained in other
countries. Research conducted in the US showed that the most common event which is a deviation from the normal
state is a fall from a height. From 3124 investigated fatal accidents, 1114 were caused by falls from a height [6].
The results of research of 2433 accidents at work that occurred in the years 1998-2003 in the Netherlands also show
that the most common reason of an accident was a fall from a height. In the analyzed period of time, 576 accidents
were caused by falling off a ceiling floor, 425 accidents were caused by falling from a ladder while 338 accidents
were caused by falling from scaffolding [7].

Fig. 4. Simulation of an accident process.

362 Bożena Hoła and Mariusz Szóstak / Procedia Engineering 172 (2017) 355 – 362

5. Summary

A methodology of analysing the accident rate as a dynamic discreet process has been developed. Part of the
presented methodology is the computer knowledge base on accidents at work and a computer model simulating the
process that is composed of many accidents. Each accident can be considered as a process consisting of a pre-accident
phase, the accident phase and a post-accident phase. Each accident occurs according to a different scenario. A set
of accidents that run according to different scenarios can be presented in the form of a directed graph.
A computer model simulating an accident process was created and calculations, which included 130 accidents
at work that occurred in the construction industry, were carried out. As a result, the critical path and therefore the
scenario with the highest probability of occurrence was determined. The conducted analyses showed that the most
accident causing area of activity in the Polish construction industry is the construction of new buildings. Carried out
simulations proved that accidents mainly occurred during the movement of employees along a flat area and during the
going to or coming back from another level. These accidents included slips, trips and falls on the same level
or to a lower level and resulted in serious body injuries or the death of the victim.
The results of the conducted research may justify the directions of preventive measures which are carried out
to reduce the number of accidents at work in the construction industry, and their implementation into construction
practice will contribute to raising the level of occupational safety in the Polish construction industry.

References

[1] Accidents at work in 2014. Central Statistical Office, Warsaw 2015.

[2] B. Hoła, M. Szóstak, Model of accident situation development in the construction industry, Czasopismo Techniczne. Budownictwo, Technical
Transactions. Civil Engineering (111), issue 1-B (5) (2014) 239–246.
[3] European Statistics on Accident at Work (ESAW) Summary methodology. Eurostat Methodologies & Working papers, European Union, 2013.
[4] B. Hoła, M. Szóstak, Analysis of the Development of Accident Situations in the Construction Industry, Procedia Engineering, XXIII R-S-P
Seminar, Theoretical Foundation of Civil Engineering (23RSP) (TFoCE 2014) 91 429–434.
[5] Control protocols from the archives of the State Labour Inspectorate in Wrocław, 2008–2012.
[6] Seokho Chi, Sangwon Han, Analyses of systems theory for construction accident prevention with specific reference to OSHA accident reports,
International Journal of Project Management, volume 31, Issue 7, October 2013, pp. 1027–1041.
[7] B.J.M. Ale, L.J. Bellamy, H. Baksteen, M. Damen, L.H.J. Goossens, A.R. Hale, M.Mud, J. Oh, I.A. Papazoglou, J.Y. Whiston, Accidents in
the construction industry in the Netherlands: An analysis of accident reports using Storybuilder, Reliability Engineering & System Safety 93
(2008) 1523–1533.