The overall aim of this study is to contribute to the solution of a problem currently faced by mining companies of ineffective LFI. It is undisputable fact that improved knowledge in the LFI strategy would lead to reduction in occupational injuries and fatalities among mining workers in Zambia. For this reason, this study is performed with the intention: to identify barriers to LFI from incident in the Zambian mining industry, to identify significant underlying conditions that set up the barriers to the LFI strategy, to formulate proposals for improving the LFI strategy in order to prevent recurrence of incidents and pre-empt new ones.
According to the mining accident statistics from Zambia’s Ministry of Mines and Mineral Development, the mining industry recorded 282 fatalities in the last thirteen (13) years. On average twenty (20) miners are fatally injured while executing their work every year while non-fatal reportable injuries recorded are above one hundred and fifty (150) during the same period. See Appendix 1 and 2 for details. This fatal accident rate is unacceptable and merits some pragmatic efforts to reverse the trend more especially that most of the fatal injuries are caused by the same known hazards.
The consequences of these incidents are very devastating at both family and national level. The social, human and economic cost of these incidents is immense given a high unemployment and poverty levels in the country. The consequence of the fatalities is very devastating to the surviving members of the family as they face dreadful future without their breadwinners. In some cases, so devastating that surviving innocent children fall out of school due to lack of sponsorship.
Table of Contents
Chapter One:
1. Introduction
1.1. Study Background
1.2. Problem Definition
1.3. Motivation
1.4. Research Aim
1.5. Research Questions
1.6. Structure of the Report
Chapter Two
2. Literature review
2.1 Background
2.2 Introduction
2.2.1 Applicable laws for Mining Companies in Zambia
2.2.2 Mine Accident Statistics
2.3 Incident/Accident models and investigation
2.4 Learning from Incidents
2.4.1 Case Study
Chapter Three
3. Methodology
3.1 Research design
3.2 Description of study sites
3.3 Questionnaire Development
3.4 Pilot study
3.5 Sample Size and sampling technique
3.6 Survey Questionnaire Response
Chapter Four
4. Discussion
4.1 Internal Consistency of the Surveys
4.2 In which step of the LFI process are major barriers located?
4.3 Which Steps Are Formally Organised?
4.4 The Variances Between the Properly organised LFI Process and the Performance in Practice
Chapter Five
5. Conclusion and Recommendations
5.1 Conclusion
5.2 Recommendations
References
Appendices
Appendix 1: Historical reportable Accidents and fatalities
Appendix 2: Mine Accidents Fatalities by Causes
Appendix 3: Chamber of Mines Employment Figures
Appendix 4: Zambia Mining Employment level 2000-2016
Appendix 5: Survey Questionnaire Safety Practitioners
Appendix 6: Survey Questionnaire General Staff
Appendix 7: Interview Questionnaire for Safety Superintendents
Appendix 8: Focus Group Discussions
Appendix 9: Mintab Cronbach Survey Reliability Analysis
Appendix 10: Survey Questionnnaire particpants responses-
Appendix 11 Access permission to study approval
Appendix 12: Ethics application Form
List of Tables
2.1 Fatalities by Accident causes in Zambian Mining Industry
2.2 Zambia Mining Safety Accidents Statistics
3.1 Number of Survey participants and their response rate
4.1 Proportions of respondents of how LFI steps is organized in Mining industry..
4.2 Causes and conditions for the main barriers in LFI process 34-
4.2 Formal Organisation of Steps (Proportions of Respondents from Mining Companies
List of Figures
2.1 Fatalities by accidents causes in percentage in Zambian Mining Industry
2.2 Trends in reportable and fatal accidents in Zambian Mining Industry
2.3 Heinrich Domino Theory
2.4 Swiss Cheese model (Reasons 1990)
2.5 Incident Causation Analysis Method Flow Chart
2.6 Learning from Incident Process Cycle
4.1 Learning from incident steps in which main barriers are located
4.2 Comparison of Steps that are organized and performed well
4.3 Overall comparison in responses per step on what is organised and what is performed well
Nomenclature
1. Fall of ground: Sliding or falling of rock from the side wall or roof of the underground excavation
2. Detonation of Explosives related fatalities: Fatal Injuries caused by concussive forces and flying rocks due to detonation of explosives
3. Mobile equipment related fatalities: Fatalities arising from risks related to motion of mobile equipment and interaction with employees.
List of Acronyms and Abbreviations
Abbildung in dieser Leseprobe nicht enthalten
Abstract
According to Mine Safety Department (MSD Reports), Mining companies in Zambia have continued to register a high number of reportable injury and fatality rates, on average about 20 mineworkers are fatally injured while executing their duties annually while non-fatal reportable injuries recorded are above 150 annually. Investigations into the causes of these accidents identify the same causal agents such as fall of ground (rock fall), blasting related accidents due to incompetent persons being licensed to conduct blasting operations, interaction with mobile equipment, underground fires among other causes. This trend has continued despite significant safety improvement efforts that have been adopted such as sharing the key learnings from incidents (LFI) for internal and external sources. However to reverse the trend, more efforts needs to be employed, starting with the identifying of the barriers that hinder effective learning from incidents and get a strong in-depth knowledge of the enabling drivers that promote harnessing of the learning potential from the LFI process.
This study is aimed at identifying barriers (bottlenecks) in the learning from incident process and their underlying causes within the Zambian mining industry. The study was conducted using both a qualitative and quantitative approach using survey questionnaires, semi interviews and focus group discussions at the three large mining companies in Zambia which had safety related datasets for the research.
The Learning from incident process is broken down into eleven (11) steps, which stand as a measure of how the Zambian Mining industry learns from past incidents. These eleven steps are incident reporting, incidence registration, investigation determination, incident fact-findings, incident fact analysis, and action formulation, prioritization of actions, communication, resource mobilization and lastly, evaluation and implementation of action plans.
The results of this study suggest that the major barriers to LFI were more prominent in incident reporting, determining of investigation, incident facts findings, incident facts analysis and evaluation of implementation steps of LFI process. The underlying cause to the main barriers identified in LFI process include a weak safety management system that lacks a proper framework to integrate the LFI and a robust audit on the performance of the whole LFI process.
Acknowledgment
To undertake this long distance programme study and combined it with formal work has been very taxing and as such needs many sacrifices from many people. Therefore, I would like to thank my wife Lydia, my sons Muma, Mutale, Bwalya and Mapalo for their resilience and bearing with me for three years of my study in which I had to spend time away from them after work and during the weekends. It would not have been easy to complete this programme without their unwavering sacrifice and support considering that the study was self-sponsored using funds from my monthly income.
I also wish to thank Mr. Barry Baker, my project supervisor for his guidance throughout my study and my parents Joseph & Rosemary.
I would also wish to thank the participants for taking part in this project and enable me to obtain the necessary information required for the project data.
Lastly, I wish to thank my supervisors at place of work and friends for the moral support rendered to me to during the course of the study programme.
Chapter One:
1. Introduction
1.1. Study Background
Mining is a very important economic activity as it provides the raw materials required for global industrialization. Due to the increased demand for metals especially by emerging economies such as China and India, there has been an increase in mining activities globally (Nicolas and coworker , 2014). Developing countries are usually target destinations for mineral exploitation owing to the low cost of production, favourable taxation and weak or lack of occupational health and safety policies for holding erring mining companies accountable, (Zambia's Auditor General Report, 2015,Vingard & co-workers,2013, 2011, Fraser and Larmer ,2010). Zambia, a developing country situated in Sub-Saharan Africa is no exception as it is one of the preferred destinations by international mining companies due to its rich mineral wealth, low mining tax and peaceful environment. The country hosts approximately 40 percent and 10 percent of the world's known copper and cobalt reserves respectively (Lydall and Auchterlonie, 2011).
The mining industry in Zambia accounts for 12% of the country's Gross Domestic Product (GDP), 70% of aggregate export value and 80 percent of Zambia's recent Foreign Direct Investment (Aguirre, 2016, Mabenge,. 2016, World Bank Country Diagnostic Report 2018;). Therefore, the sector remains a substantial source of the country's foreign exchange income and formal employment. According to the Zambia Chamber of Mines (ZCM), which is a registered alliance for mining companies, the current formal employment levels in the Zambian mining industry stand at 62236, this present 21 percent of formal private sector employment in Zambia (Zambia Chamber of Mines, 2019, World Bank Group Report, 2015). Notwithstanding the contributions to the national revenue and economic growth, mining industry in Zambia has constantly been blamed for the unreasonable rates of fatalities and lost time injuries ranging from permanent and non-permanent disabilities;
On average one hundred and fifty (150) employees are reported injured every year while a fatality injury rate is considerably high as twenty workers are fatally injured every year (MSD Reports 2018 and Auditor General Report 2015).
Sadly, many incidents recorded by the mining companies tend to be repeat incidents, which would have been avoided if a robust learning from incident (hereafter referred to as LFI) process were in place. These rates of occupational injuries and fatalities are disquieting and need urgent pragmatic actions from all concerned stakeholders (Mine Operators and Mining laws Regulators) in order to reverse the trend.
There are many proofs in demonstrating mining sector as a hazardous industry; many studies have asserted that the mining industry is one of the most high-risk sectors due to its inherent characteristics such as high humidity, unstable underground strata, mobile equipment, heavy tools and physically demanding activities (MacNeill, 2008, Elgstrand and Vingard, 2013).
1.2. Problem Definition
According to the mining accident statistics from Zambia's Ministry of Mines and Mineral Development, the mining industry recorded 282 fatalities in the last thirteen (13) years. On average twenty (20) miners are fatally injured while executing their work every year while non-fatal reportable injuries recorded are above one hundred and fifty (150) during the same period (MSD reports, 2017). See Appendix 1 and 2 for details. This fatal accident rate is unacceptable and merits some pragmatic efforts to reverse the trend more especially that most of the fatal injuries are caused by the same known hazards.
The consequences of these incidents are very devastating at both family and national level. The social, human and economic cost of these incidents is immense given a high unemployment and poverty levels in the country. The consequence of the fatalities is very devastating to the surviving members of the family as they face dreadful future without their breadwinners. In some cases, so devastating that surviving innocent children fall out of school due to lack of sponsorship.
Much as the mining companies in the country have implemented safety strategies and control measures aimed at preventing or reducing occupational safety incidents, serious accidents have continued ravaging the industry. For example significant efforts have been directed to eliminating uncontrolled fall of ground, which is a major contributor to the most fatalities in the Zambian Mines (MSD Reports, 2018) by applying rock stress analysis using 3D numerical model software (MAP 3D). Other measures put in place include use of Fall of Ground light detectors Trigger Action Response Plan (TARP) and use of remote controlled underground mobile equipment to lessen man/machine interface. Another area that have seen improvement in most large mining firms is the incident database servers such as Accident and Incident Management System (AIMS), have been developed for safety information registration and sharing. If properly managed the system (AIMS), can be used for smooth running of the accident investigation process, recording the main findings of the investigation and recommended actions necessary to prevent incident recurrence as well as tracking them to completion. AIMS can also be used to track corrective actions from visible felt leadership, safety inspections and audits to enable continual improvement of the organisation.
Another area of improvement was the introduction of incident sharing platform by the Zambia Chamber of Mines, which is a registered association of mining firms. This initiative was adopted to assist members to share LFI lessons in the quest to reverse the trend of repeat incidents across the mining industry. The main essence of LFI initiative was to drive the dissemination of incident information across the mining industry in Zambia. However, despite this initiative, mining companies have still problems in bringing down the number of high consequence repeat incidents. The recurrences of incidents with similar causes in the mining industry strongly suggests a failure to learn from past incidents. The consistent number of fatalities, coupled with high incidence of lost time injuries and permanent disabling injuries sustained within the mining industry over the past 13 years poses the question of whether the Zambian Mining Industry has failed to learn from past incidents with regard to the workplace injuries' prevention.
This study tries to identify barriers to LFI within the Zambian mining industry and the general understanding of their underlying causes so that strategies to overcome the identified deficiencies can be formulated and help to prevent fatalities and serious injuries.
1.3. Motivation
The reasons that have motivated the study on the barriers to LFI in the Zambian mining industry include the following:
a) The high fatality and injury numbers across the industry are mainly caused by the same causal agents, and most are repeated incidents (MSD reports), see Appendix 1. By gaining in-depth understanding of barriers that cause this ineffective LFI process in Zambian mining industry, better approaches may be formulated to improve safety performance and save lives.
b) Public outcry
There has been a public outcry on the number of fatal mine accidents by trade unions and the community around mining towns as they feel Mining companies are preoccupied and driven by profits.
c) The Government has also shown deep concern on the high number of fatalities at the mines as indicated by the line Minister (Zambia Daily Mail 2015).
d) By carry out this research study, the author was provided with the opportunity to apply and improve knowledge gained in Safety and Risk Management Post Graduate Diploma course.
1.4. Research Aim
The overall aim of this study is to contribute to the solution of a problem currently faced by mining companies of ineffective LFI.
It is undisputable fact that improved knowledge in the LFI strategy would lead to reduction in occupational injuries and fatalities among mining workers in Zambia. For this reason, this study is performed with the intention:
- To identify barriers to LFI from incident in the Zambian mining industry.
- To identify significant underlying conditions that set up the barriers to the LFI strategy.
- Formulate proposals for improving the LFI strategy in order to prevent recurrence of incidents and pre-empt new ones.
1.5. Research Questions
The research questions that are of prime importance to LFI that this dissertation wishes to address are as follows:
1. How do Zambian Mining Industry learn from incidents?
2. What are the barriers to LFI in the Zambian Mining Industry?
3. What are the underlying factors that pose a challenge to LFI in the Zambian Mining industry?
In the context of this research study, the LFI is looked at as the aggregation of organizational experiences that allows the mining companies to extract valuable information from past safety incidents and use it to improve organisational safety performance outcomes with time. Implementing this initiative successfully is an excellent way to put into use and bring about “organisational learning cycles” as posited by Kim and Senge, (1994).
1.6. Structure of the Report
This paper is structured into five chapters to ensure a logical run of the information and the rest of the thesis as follows:
Chapter Two: Literature Review, this chapter introduces germane and existing literature on earlier works related to LFI available from mining and other industries. It also presents incident records and statistics on Zambian mining companies from MSD.
Chapter Three: Methodological Approach, this chapter discusses the method of data collection used and compile the data to provide a more detailed insight of the barriers hindering the LFI in the Zambian mining industry.
Chapter Four: Discussions, this chapter outlines and explains the overall findings of the data obtained and identifies the key findings to related significant barriers to LFI and underlying conditions that set up the barriers to the LFI strategy.
Chapter Five: Conclusions and Recommendations, the chapter summarises the main findings of the research; statements about the main contributions of the research and recommendations for future work.
Chapter Two
2. Literature review
2.1 Background.
This section reviews existing literature and theoretical work related to the topic. The author has undertaken literature review to identify what is already known about learning from occupational safety incidents. Due to limited literature on LFI in the mining industry, gathering information on experiences and commendations from other sectors where this subject has been discussed was considered necessary in this study. The literature review was conducted in support of the author's dissertation entitled: “What are the barriers to learning from incidents in the Zambian mining industry?”
The primary focus of this literature review was to identify the challenges or barriers to learning from incidents that have hindered the prevention of repeated workplace incidents in the Zambian mining industry. A collection of the journal articles from scientific database sources including Science Direct (Elsevier) and ProQuest were used for the literature review in relation to the research topic. In additional, accident statistics and other information for the study were obtained from the Mines Safety Department of Zambia.
2.2 Introduction
Prior to discussing theories on learning from incidents, it is imperative to have a clear definition of the term incident. This is necessary in order to make an explicit distinction between an incident and accident as the two terms are often confused and used interchangeably.
Occupational Safety and Health Administration (OSHA) defines an incident as "an unplanned, undesired event that adversely affects completion of a task.” Incidents vary in severity from near misses to fatal occurrences. The term incident can also be defined as an occurrence, condition, or situation arising in the course of work that resulted in or could have resulted in injuries, fatalities, illness or damage to property Under the OHSAS 18001, 2007, an internationally recognised health and safety management standard, an incident is termed as a work-related event in which an injury or ill health respective of severity or fatality occurred, or could have occurred. In the context of this dissertation, an incident is an unplanned work related event that interrupts normal operations and results in injury or fatality.
2.2.1 Applicable laws for Mining Companies in Zambia
There are a number of safety and health laws aimed at protecting safety and health and welfare of employees in the mining industry in Zambia. The laws include the Health and Safety Act 2010, Occupiers' liability Chapter 70, Mines and Minerals Development Act 2008, Mining Regulations of 1973, The Explosives Act Chapter 115 and Explosives Regulations of 1974. Others are the Common law duty of care to employees and visitors and the Environmental Management Act 2011. The mining industry in Zambia is regulated by the Ministry of Mines through Mine safety Department (MSD), which has maintained a very inclusive database of mine accident statistics in the country. This same database was used by the author to get the invaluable source of detailed information on all type of serious and fatal accidents that happened between 2005 and 2018 in this thesis.
2.2.2 Mine Accident Statistics
Tables 2.1 and 2.2 show major causal contributors of fatalities and official numbers of mine fatalities and serious injuries over a 14-year period respectively. This data from the Mines Safety Department provides overall reported mine accidents from all the mining companies in Zambia. A cursory glance at the table below reveals that, Zambian mining industry recorded the highest number of 80 fatalities in 2005; this was due to a catastrophic event where explosives detonated and fatally injured 51 employees on the spot. Thereafter, there has been nearly constant reported fatal accidents in the Zambian mining industry over the period between 2006 and 2018.
Furthermore, analysis of data from Mine Safety Department (Figure 2) suggests that the main contributors to fatalities in the mines is fall of ground, detonation of explosives and mobile equipment /human interaction which accounted for 29, 24 and 21 percent of fatality cases respectively in the period 2005-2018 as shown in the table below.
Table 2.1 Fatalities by Accident Causes. Source: MINES SAFETY DEPARTMENT DATA MANAGEMENT, (2018)
Abbildung in dieser Leseprobe nicht enthalten
Figure 2.2 Trends in the reportable and fatal accident in Zambian mining industry 2005-2018
Abbildung in dieser Leseprobe nicht enthalten
2.3 Incident/Accident models and investigation
Incident prevention is the main purpose of all health and safety management systems of which LFI approach is an all-important integral part. An effective LFI is shaped by knowledge acquired from lessons learned from past incidents. Without proper understanding of the causes of incidents, organisations' ability to design appropriate controls to avert incidents is fragile; therefore, the understanding of how incidents happen is essential. This gives an explanation why incident causation aetiology forms a strong link with LFI concept.
Though detail discussion on incident causation theory is not part of the scope of this thesis, this section will shed some light on common accident causation models as it still form an important ground for the LFI process since they provide insight on how accidents occur. According to Arboleda and Abraham (2004), accidents causation models provide fundamental principle for incident investigations.
In endeavoring to gain deep insight of the causes of incidents, a number of researchers have developed a plethora of accident causation models of how accidents happen. One of such models is sequential model, which is simple linear type (cause-effect), it expresses an accident as a sequence of events that happened in a defined order. One of the earliest accident causation sequential model known as a domino theory was developed by Herbert Heinrich (1931) .The domino theory is conceptualized in sort of five dominos, (Ancestry and social environment, employees fault, mechanical and physical hazards) which will fall sequentially one after the other if the first domino is knocked, see figure 2.3.
The theory is focused on people as the central reasons of incidents while recognizing management as responsible for preventing them through the controlling of hazards. The weakness of this theory in relation to the LFI approach lies in its application, it sets a tone for blame culture by identifying human errors as a main reason of causes of incidents and does not consider organizational factors that lead to the occurrence of incidents. Domino theory postulates that an accident is caused by a single cause even where multi causes maybe present, for this reason it remains irrelevant in the contribution to the LFI approach.
Abbildung in dieser Leseprobe nicht enthalten
Figure 2.3: The Domino Theory
To critique, a single cause accident model, Reason (2000) developed a multi causation theory, hypothesizing that for an incident to occur there are many contributory causes that adversely combine together to cause the adverse event. The theory postulates that incidents are caused by the concatenation of multiple factors where each factor contribute to create a trajectory for an incident in a different way. This model is referred to as systemic accident model also known as Swiss cheese model, see figure
2.4 below. The multiple factors referred to include, individual, organizational and environmental factors. According to the Swiss cheese model, an organisation's defenses against failure are fashioned as a series of layers represented by slices of cheese. The holes in the slices depict weakness in individual parts of the system and show a continuous variation in size and position across the slices. Failures within the system are generated when the holes in the slices shortly align to create a hazard pathway such that a hazard goes through holes in all the slices resulting into failure. This model encompasses immediate factors arising from the operational level and latent factors on the systemic level, where the systemic factors are the deeper or underlying reasons for the operational factors as illustrated in the figure 2.4 below.
Abbildung in dieser Leseprobe nicht enthalten
Figure 2.4: Swiss cheese Model (Reason 1990)
The Swiss cheese accident causation model has been adopted by most Zambian Mining companies and use it as an incident investigation tool known as Incident Cause Analysis Method, (ICAM). The ICAM investigation tool is designed to identify and analyse the active failures that contribute to the incident and the latent hazards within the system or the organisation. The model is different from the Domino theory because it avoids restricting the investigation to the errors and violations of individuals at the sharp end. ICAM provides the ability to analyze incident information, to identify what exactly went wrong, and to make corrective actions to prevent recurrence. It is aimed at building error- tolerant defence controls against future incidents.
Accordingly, the development of this model was to allow for structural improvement of safety through the analysis of incidents causation factors and report the findings in order to share key learnings. Thus, the model is appropriately suitable tool to unravel system deficiencies, provide opportunity to understand what went wrong, fix them, and overall improve safety operations. Therefore, it is very relevant to effective LFI.
Figure 2.5 ICAM flowchart. Courtesy of BHP Billiton, (2005)
2.4 Learning from Incidents
Driving LFI process in a holistic manner can lower number of incidents and develop organizational learning culture. This outcome can be attained by incorporating the LFI into the organisation's systems leading to change in design and individual behaviours. However, just like the saying goes that it is easier said than done, effective LFI calls for defining and evaluating of all steps in the process from incident reporting to close out of the remedial corrective actions (Le Coze, 2008).
Most studies have regarded LFI as being an important element in safety management; however, in spite of its importance it is associated with a horde of challenges that limit its effectiveness (Kjellén, 2000). Some examples of those limiting factors include; the extent of employees' motivation and willingness to report incidents, Inadequate system integration (LFI not entrenched in other safety management system elements).
Other studies conducted in non-mining sector have also identified causes of why organizations fail to learn well from incidents as the under reporting of incidents. In a study by Mascini (1998) in the health care industry, he described incident under reporting as a barrier to LFI. The underlying cause to this blockage according to the study is that employees' rapport with their co-workers is taken into account when reporting incident involving the acquaintances. This behaviour negatively affects the number of incidents reported.
Furthermore, a study by Sanne, (2008) acknowledged that one of the most important aspect of LFI that can prevent incidents is the Incident-reporting schemes. The study however identified the aspect of under reporting as a main barrier to LFI in the railway industry, this was attributed to failing to integrate the incident-reporting scheme within the organizations. From the findings of the above studies, it can be strongly deduced that step of incident reporting if not properly managed can have deleterious effect on the entire LFI process because the necessary information required in identifying the root causes of the incidents is not made available.
In contrast, the Zambian mining companies are compelled by law to report incidents to MSD and therefore the mining industry is not short of information to form the much- needed input in the LFI process. However, what remain to be unravelled is the understanding of why the industry is failing to curb repeat incidents by using the available information?
Other authors on the subject of LFI have argued that organization would maximize lessons learned from incidents by analysing them in a social context(Lukic and coworkers 2010).This means that, taking a robust approach in identifying causes of failures and address contributing factors that influence the existence of identified deviations.
Some authors have posited that the incident precursors provide a window of improvement and contribute to the effectiveness of LFI process. They have argued that incidents are in most cases preceded by an incubation stage where latent deviations are accrued until they are triggered by active failures (Reason, 1997; Turner 1978; Pidgeon and O'Leary 2000). In actuality latent deviation in this concept, include management decision taken without factoring in safety risks, failure to take decisive actions to correct high frequency low consequence. However to achieve good safety performance organization must take a robust approach in identifying causes of failures and address contributing factors that influence the existence of identified deviations in specific incidents rather than focusing on the active failures.
However, this approach in reality is not properly addressed in the Zambian mining industry as incident investigation process mainly focuses on the direct causes of incident which tend to incline towards identifying the specific failures in the specific incident rather than finding contributing factors at the organization level. This means that most “underlying causes” of the incidents are mostly deduced from analysis of direct causes rather than in-depth analysis of organizational factors that allowed deviations to exist in the first place. A very good example is the case study from one of the case companies below:
2.4.1 Case Study
Short Description of Incidents:
On 15th April 2014, two underground rig operators were found unconscious in a seating position leaning against the water column near a drill rig at a working level underground by a supervisor. There were no visible indication of physical injuries however, the postmortem reviewed that the cause of death was electrocution
When the incident happened the eye witnesses did not report the matter immediately to the supervisors but instead disturbed the scene of accident and reported the matter as a case of heat exhaustion.
During the fact analysis, it was established that the two deceased did not die from heat exhaustion related conditions contrary to what was earlier reported. It was also established that the Boomer Drill Rig was connected to the faulty electrical source. What happened was that when the power was turned on, the boomer rig was live because the live phase and earth was swapped around at the electrical box. And in additional, the electrical box was not earthed as a result of this there was no completed circuit and the earth leakage could not pick up any mal functioning or that the wires were not correctly connected. The boomer rig was live because the live wires from the boxes were directly connected to the earth of the boomer however, it could not complete the circuit because of the large rubber tyres. When the duo touched the live machine drill rig, they completed the circuit to earth and 550 volts went through them. The contracting company distorted the incident description. The contractor underground manager conducted the cover up at a high level. The drill rig was removed from the scene of incident and the cable was disconnected at the substation and boomer box to hide the evidence, the eyewitnesses all stated that the two deceased had succumbed to heat stroke and set up the scene to look like a heat related incident. All four eye witnesses who claimed the incident was heat related had undergone the polygraphic examination and failed. The following were concluded as the direct causes and contributing factors:
- The incident was reported only after the contractor supervisor had arrived on site
- The incident was reported incorrectly as heat exhaustion
- Based on the result of the postmortem the cause of death was electrocution
- Unauthorized personnel did a substandard job on the connection of the boomer box, normal procedure requires the mine appointed engineer to carry out isolation and connection on any equipment and not a contractor personnel.
- Contractor exceeded his authorization
- The earth leakage device did not pick up a faulty completed circuit and cut the power.
Two years later on 3rd August 2016, Three employees were electrocuted while they were installing a new B2400, 550V Pump at a water sump underground. As part of the installation process, a cable joint had to be made. Upon completion of the cable joint, a contracting company electrician left the area, proceeded to the main substation approximately 230m away, and switch on the power. The investigation into the incident revealed that, there was an electrical fault on the pump starter circuit and the earth leakage protection inside the pump starter panel failed to trip the power. It also stated that the pump was energized by a contractor electrician while the pump crew members were still standing in the water, a strong violation of energy isolation and lockout procedure.
From the above two incidents, it appears that when an incident occurs, the causes leading to its occurrence remain the target of investigation rather than exploring why the system allowed the existence of unsafe conditions and acts to occur in the first place. There are noteworthy resemblances of underlying causes and organisational factors between the two fatal accidents that were not addressed in the first incident hence potential learnings were lost. If proper investigation and all steps of LFI were performed well from the first incident, the second incident would have been avoided. This suggest a problem in the LFI process and merit a study to understand the barriers that hinder learning in the Zambian mining industry and their associated underlying causes.
Therefore, the author's motivation for conducting this study dwells on the fact that by identifying the barriers that hinder the learning from past incidents and find counter measures to improve the LFI process model, future incidents could be avoided. To ameliorate learning from incidents mining companies in Zambia need to identify problems that is making it difficult to learn from past incidents and address them, implying that detailed knowledge of what caused the difficulties in learning from incidents needed to be gathered. The study will therefore prove a valuable piece of inquiry that may benefit mining companies in Zambia.
While there has been a number of research publications on the subject of learning from incidents in other industries, there is a dearth of literature on the barriers to the learning from incidents in the mining industry and this dissertation addresses this gap by taking a practical approach to the study of leaning from incidents. There are numerous LFI models that have been designed to assess and analyse the effectivity of the organisations' learning from incidents process (Drupsteen, Groeneweg, & Zwetsloot, 2013; Jacobsson, Ek, & Akselsson, 2011; Lindberg et al., 2010; van der Schaaf, 1992 such as one from Energy Institute, Glasgow Caledonian University LFI and TNO LFI analytical framework.
The study used the LFI analytical framework developed by TNO and Drupsteen (2014) to investigate how the mining companies learn from incidents and where the possible barriers are located in the learning process model. The incident-learning model consists of eleven (11) steps and is shown in Figure 2.6. This in actuality means that the LFI process is kind of a sequence of steps, where the quality of the succeeding step is dependent on the preceding one. Simply put that poor quality input inevitably produces faulty output, for example the remedial actions the investigation team presents would not be effective if the underlying causes were not identify. Lindeberg (2010) also stressed the importance of follow up steps in the entire LFI process to ensure effective remedial actions.
The model represents learning from incidents as a whole process, from incident reporting, gathering of information on incidents and analysing that information, to the use of key learning lessons for the enhancement and the evaluation of effectiveness of the remedial control measures taken.
The fact that repeat incident keep on reoccurring, strongly suggest that the learning process has barriers that can be found in each of the eleven steps.
The first two steps of incident reporting and registration is aimed at identifying that a specific situation has occurred and presents opportunity to learn from it. Thus incident reporting forms the necessary stimulus for the learning process to kick-start. For this reason, some form of incident notification system is required (Armitage, Newell, and Wright 2007; D'Souza, Koller, Ng, et al. 2004) as well as a conducive working environment (just culture) ought to be present ( Hopkins 2006, Firth-Cozens 2004).
The reporting of incidents is vital for the complete learning process to take place, for instance if an incident is concealed, no learning lessons can be drawn from it and thus remedial actions will not be made (Drupsteen and Hasle.2014).
According to Argyris and Schön (1996), all learning begins with the gathering of information, which is the learning input. The importance of information in occupational health and safety can not be overemphasised thus the reporting of incidents is a legal requirement under the Zambian laws, the Occupational Health and Safety Act No 36 of 2010, Mines, Minerals Development Act 11 of 2015 and Explosives Act No. 14 of 1974. To fulfill this legal requirement major mining companies keep their reported incidents in their respective database systems for retention.
Once the incidents are reported and registered the next steps are to determine the scope and level of the investigation, the collection of information, and analysis of the incident facts. These three steps are critical because the learning from incidents process requires a detailed interpretation of the incident causation factors, both immediate and root causes (Reason 1990), to offer a chance of reducing likelihood of reoccurence. These steps provides strong perceptive into the incident causation factors and thus it is vital to abviate hindsight as far as possible (Brazier A, 2018 ).
Premised on the outcome of this investigation, recommendations for safety improvement formulated and key findings shared both internally and externally across the industry in the form of incident learning alert. The incident key learning alerts, assist workers to understand how to avoid similar incidents from reoccurring. The acquired knowledge would be converted into action to correct the deficiencies.
The planning of corrective actions steps in the LFI cycle , includes the prioritisation of the actions, generation of recommendations and the drawing up of a realistic action plan that has to be cost effecive,specific measuarable attainable relevant and time specicific. (Bhimavarapu and Doerr 2009).
The last three steps of the learning from incident cycle include communication of the remedial action plan to the people assigned to execute the actions, and to all concerned stakeholders to promote ownership of the assigned tasks (Barret, Haslam, Lee, et al. 2005). Availing resources, such as finances skilled labour and time, might be valuable to execute the actions as planned. It is worth mentioning that the remedial action plan and its goals are disseminated organization wide (Bahn 2009), principally to show management commitment to health and safety programmes and to share the "key lessons learned" from the investigation findings.
The implementation of the actual remedial action plan process should be evaluated in order to assess whether the plannned actions are executed or not and also to verify the effecual of the same actions. In an event that an action fall short of the expected outcome, the causes can be checked by applying quality management system continual improvement model of Plan-Do-Check-Act cycle (Deming, 1982) and thus forms the iterative aspect required in the “learning from incidents process cycle. The outcome of evaluation step is keystone to the refining the learning proficiency of the organisation to prevent the reoccurence of incidents. Figure2.6 below shows the schematic steps involved in the learing from incident process..
Abbildung in dieser Leseprobe nicht enthalten
Adopted from TNO
Chapter Three
3. Methodology
3.1 Research design
This study used a survey questionnaire strategy as a means of collecting data to be analysed in support of the research objectives. The survey questionnaire approach is considered the best strategy for this type of study since the author had background knowledge on the relevant factors such as the kind of information required and the target group to provide responses to the survey in order to achieve the research objectives. This strategy has been adopted following its successful usage in similar studies (Denscombe, 2019). Another aspect that prompted this strategy is the concern to meet research completion target within definite time, it would be unwise to settle for a strategy, which could not feasibly be completed with the set deadline.
A survey strategy provides for quantitative or qualitative approach depending on the nature of data to be gathered and how collection is planned (Kothari, (p 34 1990). This research utilized both qualitative and quantitative approaches in the analysis of the data. The data was collected through survey questionnaires, focus group discussions and semi structured interviews. The participant's responses to the questionnaire were collated and quantitatively analyzed by use of statistical software, Minitab 16, excel and Nvivo software.
The focus group comprised of sixty-six (66) safety officers mainly from contracting companies and six (6) Safety Superintendents drawn from different departments within the case companies. The reason for high number of contractors in the focus group was that the two incident cases that were discussed in the focus discussion involved contracting company employees. The Safety Superintendents were chosen because of their in-depth knowledge of the investigation of incidents. The aim of semistructured interviews with the Safety Superintendents and the focus group discussions was to confirm reliability and validity of responses provided in the survey questionnaires by considering another source of data.
3.2 Description of study sites
While there are several small and medium scale mining activities for various commodities in Zambia, most of the recorded mining accidents and incidents are from the major copper producers as copper mining is the mainstay of the Zambian mining industry. Therefore, this study was conducted on three mining companies Zambia on the Copperbelt and North-Western Province. These mines were chosen because they are largest mining companies, which together account for 70 per cent of copper production and have contributed to most reportable injuries and fatalities in Zambia. See appendices 2 and 3 for details.
3.3 Questionnaire Development
The design of the survey questionnaire involved extensive discussions with the country's Safety Practitioners network, the Zambia Occupational Health and Safety Association (ZOHSA). These discussions were made possible through deliberations from members of the Health and Safety Practitioners network on the online media platforms, which hosts nearly all Occupational Health and Safety Practitioners across the country that include mining, transport, manufacturing and construction industry
Two versions of structured questionnaires were developed based on the Netherlands Organisation for Applied Scientific Research (TNO) LFI process model and drawing relevant information on the Incident Causation Analysis Method (ICAM) model developed by James Reason, (1990). While safety practices and procedures are for the most part developed by Safety Practitioners, the objectives of these procedures and practices are only achievable if management rightly implement them and adhered to by the shop floor workforce. Therefore, different versions of survey questionnaires for the Safety Practitioners and employees were developed to include twelve (12) and eighteen (18) questions respectively so as to get the perspectives of both the developers of the procedures and the shop floor workforce.
Each question required a response against a four (4) point Likert Scale rating system ranging from one extreme to another, ‘strongly agree' to ‘strongly disagree' with no neutral point in order to force a choice. On the Likert Scale, (4) four (strongly agree') denoted the highest score and (1) one (strongly disagree') being the lowest. These responses were subjected to quantitative data analysis so as to get various statistical and technical parameters that can measure level of LFIs.
Furthermore, an additional structured and qualitative interview questionnaire was developed for Safety Superintendents so as to have a basis for comparison between the two methods of analysis that is, the quantitative and qualitative analysis methods. At the end of the survey questionnaire for the Safety Practitioners, there was space for the free text where the survey participants were required to specify in which of the eleven steps of LFI model process, in their view, the utmost bottleneck in their company was located.
The questions were modified several times and the questionnaire was reviewed by the Safety Practitioners from non-mining sector for clarity, completeness, and the use of appropriate terms.
3.4 Pilot study
Trial versions of an online questionnaire and printed questionnaires were distributed to a small sample of Safety Practitioners and employees to pre-test them by the author in order to check the wording and clarity of the LFI questions. The objective of the pilot study was to evaluate how easily the target user groups would understand the questions.
The author chose nine (9) Safety Practitioners from the three major mines and five (5) miners from a local mine who voluntarily elected to take part in the pilot study. The nine (9) Safety Practitioners and five miners received online and printed versions of the LFI model questionnaire respectively and were requested to comment on the limpidity, appropriate terminologies and completeness of the questions. On both versions of the questionnaire, an additional free writing space was provided which asked for the total time required to complete the questionnaire, clarity assessment, and any other comments of concern.
All fourteen (14) respondents completed the questionnaires of this pilot study. The average time required for the two groups to complete the questionnaire was thirty minutes. Eight (8) respondents raised concerns regarding the limpidity of some questions. Premised on the response, some questions were amended to address and improve on the clarity of the questionnaire.
3.5 Sample Size and sampling technique
The sample consisted of sixty-five (65) Safety Practitioners who were selected purposively from Mining companies. The Safety Practitioners were members of the Zambia Occupational Health and Safety Association (ZOHSA), a network of Safety Practitioners in Zambia. The sample size consisted of the ZOHSA active members from the three mining companies and were easily accessible through the Association group email address. The total head count of Safety Practitioners from the three mining companies under the study was seventy (70) but five (5) were not members of ZOHSA. As special knowledge in the incident management and LFI and related issues was required from the respondents, the application of purposive sampling in this study was very appropriate. As suggested by the terminology, purposive sampling is used to choose a sample for a particular purpose. Safety Practitioners were considered in the survey as they are more conversant with the actual LFIs process in their Companies and as such, they were better placed to offer true assessment on both actual and formal LFIs process.
Considering the limited time and geographical distance of locations of the mines, an online self-completion questionnaire method was the most appropriate method of getting a reliable sample of the hard-to-reach Safety Practitioners.
The final versions of the questionnaire were distributed to the selected participants using a ZOHSA group email address through a link of the e survey tool while printed questionnaire copies were distributed to the selected local mine employees through their supervisors. Collection of completed questionnaire from the local mine employees was relatively easy since the author of this thesis works for the same mining company.
3.6 Survey Questionnaire Response
The overall response rate of 72% from both Safety Practitioners and local mine employees was achieved which is a very good response rate for this survey, considering, a higher share of hard to reach respondents e.g. (The Safety Practitioners due to geographical locations of the mine) see table below for number of participations by target group.
Table 3.1 showing number of survey participants and response rate
Abbildung in dieser Leseprobe nicht enthalten
Furthermore, as part of this study, focus groups from one of the mining firms were put up with the aim of identifying the steps in the LFI process where the barriers to LFI are located as well as identifying their underlying causes and conditions. This approach provided the much required triangulation to some responses from the survey. Triangulation is the practice of validating the outcomes of a study through two or more methods to check its consistency and validity. This was done in one session of four hours allocated to each group for deliberations and another twenty minutes of sharing the findings. The focus groups study centered on incident review in order to provide information on how LFIs was formally structured and organised and how it is performed in daily practice in the Mining industry. And it also focused discussing LFI steps where the main barriers are located based on the incidents provided in case study and the underlying causes of these barriers to the LFI process
The focus groups reviewed the information on incident reports, incident investigation procedure and the detailed investigation of reports for two incidents and any documentation related to evaluation of close out actions generated from the two incidents to enable the research to determine which step was properly organised or not. The six focus groups constituted of eleven (11) participants each and these were drawn from different departments including, underground, open pits, engineering services and processing plants. Two similar incidents that occurred between April 2014 and August 2016 and their full investigation reports were presented to the focus group for reviewing and discussion during the study and address the questions below: See appendix 8 for details.
“In what steps are main barriers to learning identified (i.e. bottlenecks) according to the LFI process model?
What are the underlying causes and conditions for the challenges to LFIs?
The use of focus groups provided for a wide range of perspectives and the interactions amongst group participants and served as a way of helping people to come up with ideas and deliberate on the causes for ineffective learning. This approach was necessary because it enabled the group to identify the underlying causes of the difficulties in the individual organizational learning process in order to improve the entire LFI steps.
The focus groups were guided to centre their discussion on how they had learned from the two incidents and the challenges faced in acquiring the knowledge rather than discussing in depth the individual incidents presented in the questionnaires.
Since the focus group study was aimed at deciphering specific situations such as identifying the challenges in learning and underlying causes of these difficulties, information validity was vital and as such, four independent Safety Managers reviewed the data for a second unbiased opinion in order to eliminate the bias of the researcher. Research bias refers to the effect of the values and theories of the researchers to influence the results, in order to represent a certain result.
The deliberation transcripts from each focus group were compiled in report form and were analysed to assess the LFI process steps in which main bottleneck were located and to identify the underlying causes of the same bottlenecks that lead to ineffective learning. The LFI steps provided the primary set of codes. All compiled information from the six focus groups were classified in relation to the LFI process steps thereof. By open coding the information for each LFI process step resulted into the final classification, thereof representing causes of the difficulties in LFIs. See table 4.1 in Chapter Four (4).
The correlated results from focus group discussion helped in development of the semi structured interview questions based on the steps of LFI process model. The semistructured interviews with questions such as “In your own understanding, in what step of the LFI process do you think most difficulties to learning arise?” including one asking about the most successful step in the LFI process among others. The interviews were conducted with the six Safety Superintendents who headed the focus groups.
Chapter Four
4. Discussion
This chapter discusses the results of the survey questionnaires that were circulated to the Safety Practitioners and members of staff across the Zambian mining industry. The survey conducted amongst Safety Practitioners considered three facets bordering on the quality of how various learning from incidents steps are organized, the related performance of the respective steps in daily practice and at what point in the learning from incidents process is utmost learning potential lost? Additionally, the survey that targeted the general workforce addressed the individual perceptions on the incidents learning processes and the various factors involved. While the findings from the focus group discussions and semi structure interviews, build on the data from the survey questionnaires for consistency checks.
4.1 Internal Consistency of the Surveys
This study ensured that internal consistency, that is, how closely related the set of responses from the survey questionnaires were as a group. This was achieved through the determination of Cronbach's alpha using a statistical software, Minitab 16. Cronbach's alpha is sometimes referred to as the scale of reliability. Some scholars have technically argued that Cronbach's alpha is not a statistical test but rather a coefficient of reliability or consistency. Cronbach's alpha (a) is mathematically considered as a function of the number of test items and the average inter-correlation among the items. The formula for Cronbach's alpha is shown below for conceptual purposes:
Abbildung in dieser Leseprobe nicht enthalten
[...]
-
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X.