Learning Embedded in Structures:  
Facilitating Systems for Learning  
within Offshore Drilling Operations 
Trygve J. Steiro 
Norwegian University of Science and Technology, Norway 
Trygve.j.steiro@ntnu.no 
Tor Erik Evjemo 
SINTEF Digital, Norway 
Torerik.evjemo@sintef.no 
Purpose: The current study examines the contribution of formalised planning to 
ensure success in drilling operations seen in the light of organizational learning theory.  
Methodology: The applied methodology is observational fieldwork including 
interviews, document studies and informal talks on a drilling rig on the Norwegian 
continental shelf conducted by the authors lasting for 6 days including 120 hours of 
observational fieldwork.  
Findings: It is described how procedures are adapted to the specific context in which 
a job is performed and how the adaption of procedures is integrated into the formalised 
organisational learning processes.  
Originality: Organizational learning is seen through praxis and the dialectical 
relationship between formalised work and work as done, that is, the formal 
structures involved when planning work and how procedures materialise during 
actual work, and the relationship to learning per se.  
Keywords: organizational learning, formalised planning, drilling operations  
Introduction 
The Normal Accident perspective (Perrow, 1984) claims that the degree to which technology 
is characterised by tight coupling and/or complex interactions have implications for the need 
for centralised versus decentralised control of the operations. A system is tightly coupled if 
disturbances propagate rapidly throughout the system, and there is little natural slack or 
redundancy that allows people to improvise ways to contain the disturbances. Complex 
interactions cause systems to behave in unexpected ways and may catch personnel by surprise. 
A system with high interactive complexity can be effectively controlled only by a decentralised 
organisation; for example, highly interactive technologies generate many non-routine tasks. 
Such tasks are difficult to program or standardise. Therefore, the organisation has to provide 
operators at the sharp end considerable discretion and encourage direct interaction between 
them. On the other hand, a system with tight couplings can be effectively controlled only by a 
highly centralised organisational structure. If disturbances appear, a fast and coordinated 
response is required, and this calls for centralisation. The current study is rooted in the tension 
between interactive complexity that requires decentralised organisation and close links that 
require centralisation, that is, planning from onshore that are to be executed in operations 
offshore.  
Another aspect is that parallel activities may cause conflict, so strictly centralised coordination 
is needed to avoid conflicts between activities. Rosness et al. (2010) point out that there is a 
dilemma if a system is characterised by high interactive complexity and tight couplings 
(Rosness et al., 2010:49). According to Perrow’s (1984) logic, the organisation needs to be both 
centralised and decentralised. This paper examines aspects of how this dilemma is handled on 
an offshore drilling rig, taking the application of planning per se, including procedures, as a 
International Journal of Management, Knowledge and Learning | ISSN 2232-5697
                    Volume 10 (2021) 1–20 | https://www.doi.org/10.53615/2232-5697.10.1-20
 
 
 
starting point seen in the framework of learning. Ose and Steiro (2020) studied the planning 
processes of drilling operations and found learning to be a significant enabling factor for 
ensuring successful operations.  
In a study by Bourrier (1998), two French and two US nuclear power plants were studied, 
showing how formal procedures frequently needed to be adjusted. However, the processes in 
the French nuclear power plants were quite different from the US nuclear power plants. For 
example, the French personnel adjusted the procedures on the floor without the management 
knowing—the maintenance personnel noted and shared the adjustment in a notebook. In the 
two US nuclear power plants, there was more verbatim compliance. When challenges were 
faced, they informed the foreman and asked for a new work order when the task was unclear. 
In the US plants, the middle management with the formal authority to authorise modifications 
of procedures was much more available. In the French, on the other side, they were less 
available due to staff reductions in middle management. Consequently, secret practices and 
pockets of private information developed, and the organisation as a whole failed to learn from 
successful adaptations. However, the US power plants could be termed self-correcting 
organisations because they had mechanisms such as availability and close cooperation to 
reconcile prescribed work and ‘work as done,’ and to make the knowledge of the workers 
available to the greater organisation (Bourrier, 1998). Both the two French plants and the two 
US plants could be termed machines from the perspective of Morgan’s (1980) metaphors on 
organisations. However, the US plants had clearly stronger evidence of brain attributes, 
demonstrating the metaphors as perspectives of organisations and not seen as mutually 
exclusive, but rather that ‘different metaphors can constitute and capture the nature of 
organisational life in different ways, each generating powerful, distinctive, but essentially 
partial kinds of insight (Morgan, 1980: 612).  
Offshore drilling operations are considered a high-risk activity demanding centralised control 
in line with (Perrow, 1984) to ensure safe operations. At the same time, crews may confront 
considerable interactive complexity in, for example, drilling operations. We assumed that the 
combination of tight coupling and interactive complexity might create operating and learning 
challenges similar to those facing a nuclear power plant, albeit less severe. In particular, the 
need for centralised control could lead to a strong reliance on procedures, whereas the 
interactive complexity could create situations where the procedures could not be readily 
applied. We wanted to explore the means to create a self-correcting organisation under these 
circumstances within a drilling operational context. 
Offshore life has been considered a ‘different world’ (Hellesøy, 1985). The crews operate on a 
2/4 schedule, meaning that they are two weeks offshore at work and then four weeks at home 
(onshore). The crews typically start on a night shift rotating after one week to the day shift 
(Hellesøy, 1985). Sætren and Laumann (2014) point out that they work a 12-hour shift every 
day, and they usually share all the meals together. This means they have a good opportunity to 
get to know each other quite well. The ‘vacant period’ between working hours can also be 
subject to company courses and training. In total, six crews are needed to operate each rig. 
Offshore installation managers (OIMs) are on duty 24/7, meaning that there are three OIMs on 
each rig. The same holds for the safety officer. This means that handover, knowledge, and 
experience transfer are a challenge and very interesting from the perspective of organisational 
learning. But it also means that it takes time before the knowledge is distributed and that there 
might be a time and context challenge when transferring experience.   
This paper focuses on drilling operations by exploring the implementation of formalised plans 
per se, for example, the drilling plan, including procedures and guidelines concerning 
organisational learning on-board the drilling rig. 
Steiro & Evjemo | Learning Embedded in Structures 2
 
 
 
The following research questions will be answered: 
• How is learning structured through formalised planning? 
• How can actual learning be conceptualised to improve our understanding of and 
strengthen applied organisational learning on a drilling rig? 
Theoretical considerations 
The role of Plan-Do-Study-Act (PDSA) within a machine bureaucracy 
According to Mintzberg (1979), a machine bureaucracy can be defined as a management 
structure with a high degree of formalisation and specialisation. In a machine bureaucracy, 
decisions are made at the top level and mechanically carried out at the lower levels. The design 
of a machine bureaucracy tends to be as follows: highly specialised, routine operating tasks, 
very formalised procedures in the operating core, a proliferation of rules, regulations, and 
formalised communication, large-sized units at the operational level, reliance on the functional 
basis for grouping tasks, relatively centralised power for decision-making, and an elaborate 
administrative structure with sharp distinctions between line and staff (Mintzberg, 1979). We 
see a similar organisational understanding in Morgan (2006) through his machine metaphor, 
where precision, efficiency, and predictability are key characteristics based on a bureaucratic 
and functional specialisation of tasks and responsibilities. 
The organisation of drilling operations on-board an offshore drilling rig shares many of these 
characteristics. The rationale for this is that drilling companies operate in rather stable 
conditions with a less dynamic environment (Burns & Stalker, 1961; 1960). Mintzberg’s (1979) 
classification can be viewed as an overall typology. However, during drilling operations, 
unforeseen conditions do occur, and more flexible patterns of organising are needed, in line 
with LaPorte and Consolini (1991) and being able to adapt and execute double-loop learning 
(Argyris & Schön, 1978: 1996). The same holds for a machine bureaucracy like an aircraft 
carrier where the main concern is to decide what Standard Operating Procedures (SOP) to apply 
regarding a given situation (LaPorte & Consolini, 1991). The capability to respond effectively 
in novel and unexpected situations plays a crucial role in these processes (Simon, 1991).  
Planning is essential for machine bureaucracies like drilling organisations. For example, the 
concept of Plan, Do, Study, Act (PDSA) was invented in organisations like machine 
bureaucracies (Deming, 2000; Womack, 1990). PDSA is a quality improvement system. PDSA 
is a process that follows four steps similar to a scientific methodology (Sobek & Smalley, 2008; 
Lillrank & Kano,1989). The PDSA pillars can be explained as follows. Firstly, planning is 
initiated when a potential for improvement is identified; plans need to be developed for new 
goals to be achieved. Secondly, do is the execution of the plans. Thirdly, at the same time, data 
must be collected to be able to evaluate the effort. Therefore, a study is performed after a while 
to assess plans and routines. Does the effort lead to actual improvement? Plans and routines 
need to be refined to be more accurate. The study is about assessing whether the effects of the 
implementation are sufficiently verified (Sobek & Smalley, 2008). Fourthly, the act is the last 
phase, and when eventual changes have been made, the process continues. PDSA continues 
further. 
In our second case, the drilling entrepreneur reported that they had used a PDSA system since 
2011. Imai (1986) points to the importance of spreading the improvements. Otherwise, the 
results will remain local and not spread to a wider audience (Sobek & Smalley, 2008). Structural 
problem-solving is an important fundament in improvement initiatives. PDSA/A3 is a structural 
problem-solving method that has become a symbol for the learning process improvement efforts 
it represents (Berger, 1997; Deming, 2000). A3 is based on an A3 sheet format that covers the 
problem description and the solution visually. The idea is that a visual A3 illustration can easily 
Steiro & Evjemo | Learning Embedded in Structures 3
 
 
 
be communicated and closed when the improvement is standardised (Shook, 2008; Sobek & 
Smalley, 2008).  
A basic precondition for continuous improvement initiatives is to create an organisational 
learning environment (Bessant et al., 2001; Wu & Chen, 2006). However, programs like LEAN 
and most continuous improvement programs depend on scaffolds on standard operating 
procedures (SOP) as a central and important part of the philosophy and methodology. However, 
in several cases, these standards fail to consider how individual insight and experiences are 
transformed into action, thereby potentially limiting learning per se (March & Olsen, 1975; 
Kim, 1998). 
Organisational plans and the relation to practice 
According to Suchman’s concept of situated action (1987, 2007), plans do not determine action; 
they are merely resources for action. Suchman’s original work illustrates how users trying to 
follow expert guidance systems on photocopiers based on the idea of intelligent machines 
failed. She challenged the idea within artificial intelligence (AI) and cognitive science at the 
time that machines could be designed to understand users’ goals and thus guide humans to meet 
those goals by following a predetermined plan. Suchman (1987) showed the reasoning behind 
designing an ideal plan grounded in a presumption of compliance that did not correspond to 
how the plan was carried out in practice by the end-users themselves. Suchman (1987) argued 
for the need to understand the role of plans in everyday work, acknowledging that plans are 
static constructions while the work context is ever-changing, and we rely on interpretation and 
improvisation to solve everyday challenges. Her work has had a major impact on research 
traditions like human-computer interaction (HCI) and the more socially oriented computer-
supported cooperative work (CSCW). However, and perhaps strangely, her work is rarely 
referred to within the safety research literature, although she is one of the pioneers with regard 
to problematising the application of plans relative to actual work practices. Steiro, Thevik, & 
Albrechtsen (2017) concluded that both compliance and resilience are equally important for 
successful operations and that a certain degree of both is required.  
Hollnagel (2013) argues that the traditional view on safety, which he labels Safety-I, pays 
attention to the things that go wrong but fails to study successful performance. Thus, unwanted 
events need to be contained and kept at a minimum by minimising errors via identifying their 
causes. By designing barriers and favouring compliance regarding plans, procedures, and 
regulations, the Safety-I perspective mainly addresses work as imagined to ensure safety 
(Hollnagel et al., 2013). On the other hand, safety can be understood based on what goes well, 
for example, what key stakeholders deem as successes regarding naturally occurring work 
practice. This approach, which Hollnagel (2013) labels Safety-II, requires attention to work as 
done since it implies that people make sensible adjustments to emerging demands related to 
their particular activity at the moment. Importantly, and in conjunction with Suchman’s (1987; 
2007) original work, this view on safety acknowledges that successful work accomplishment 
does not rely entirely on strict compliance with formalised plans and procedures; informal work 
factors are as well important.  
Consequently, studying actual work is very important as input to understand why adjustments 
to standard operating procedures (SOPs) are sometimes also necessary as efficiency-
thoroughness trade-offs (ETTO) (Hollnagel, 2009) or to examine whether there is observed 
practical drift (Snook, 2002). Practical drift occurs when local practices are changed and drift 
away from the design concept, which assumes tight couplings between the different elements. 
Snook (2002) argues that this creates vulnerability if the system changes to a tightly coupled 
state, even if this state was anticipated in the design phase. In situations where tight couplings 
were assumed and necessary, there are loose couplings because of drift in practice away from 
Steiro & Evjemo | Learning Embedded in Structures 4
 
 
 
the design idea (Snook, 2002). A similar example can be seen from the Longford accidents in 
1998 (Hopkins, 2000), where practice drifted away from the design specification without being 
followed up and assessed centrally. An opposite approach is found in Bourrier´s (1998) study.  
The institutionalisation of learning and the importance of showing humaneness 
Hernes and Irgens (2012) are critical of the organisational learning literature, claiming that too 
much focus has been on the mismatch between the expectations and outcomes. This is central, 
but studying continuity is equally important. Learning under continuity, the authors claim, 
requires an investment of effort, mindfulness, and preparedness for change, even if no 
perceptible change occurs. The idea that an organisation could learn and knowledge could be 
stored over time, was a key breakthrough, which was first articulated by Cyert and March 
(1963). A significant portion of the literature on organisational learning is founded on the 
individual learning theory, while social learning theory in the organisational learning literature 
has grown out of criticism of the individual approach regarding learning (Brandi & Elkjaer, 
2011). This individual learning contributes to a transformation process in the organisation 
(Pedler, Burgoyne, & Boydell, 1990). Thus, organisations should adopt flat, decentralised 
organisational structures that facilitate open communication and dialogue (Pedler, Burgoyne, 
& Boydell, 1999). Crossan, Lane, & White (1999) also argue for a dynamic approach to 
organisational learning. This is in line with Huber (1991), Nonaka (1994,) & Kim (1998). 
Crossan et al. (1999) further argue that, through interpreting, a group can share understanding 
and provide common meaning to phenomena and incidents. Here there is an integration both of 
concepts and practices. Institutionalising means that learning is embedded in plans, systems, 
and strategies which can be termed an organisation’s infrastructure, and our focus is on 
institutionalised learning through practice. Jerez-Gomez et al. (2005) mention four dimensions 
regarding organisational learning ability. These are engagement from the management, system 
perspective, openness, and experimentation. Another aspect is procedural memory that refers 
to the stored experience, knowledge, and skills reproduced on a routine basis by organisational 
members (Cohen, 1991; Moorman and Miner, 1998; Singley & Anderson, 1989). Morgan and 
Liker (2006) claim that the ability to learn through continual improvement is probably one of 
the most powerful approaches to competitiveness. Learning is also important in systems with 
the need to adjust and make sure that the organisation is not ‘drifting into failure’ (Snook, 2002).  
Bolman and Deal (1989) observed early that total quality management (TQM) systems–of 
which PDSA is a part—are good examples of how the structural and human relations frames 
are complementary and that the combination is the precondition for the success of such systems. 
Jerez-Gomez et al. (2005) claim that openness and freedom to experiment are important for an 
organisation to achieve and maintain organisational learning ability. The dimension 
experimentation is linked to the employees’ freedom to search for new ways to solve tasks, as 
is the freedom to take the experimental risk (Goh & Richards, 1997; Jerez-Gomez et al., 2005). 
However, the authors may or may not have included high-risk industries, so the experimental 
risk should be approached carefully within this context. Huber (1991) commented early that the 
distribution of information influences the degree of organisational learning. If more people have 
access to information, there will be more and different interpretations of the same information. 
De Geus (1988) points out that scenario building enables multiple scenarios to be developed. 
But more importantly, it enables conversations between people inside the company and 
outsiders of the company that can be viewed as driving forces of a social, political, or 
technological nature. We argue that De Geus’s (1998) findings can open up and invite for more 
double-loop learning (Argyris & Schõn, 1978; 1996) and more focus on the trade-off between 
exploitation and exploration (March, 1991). Von Krogh, Ichijo, and Nonaka (2000) point out 
that the key ‘quality of knowledge workers is their humanness’ (2000:12). The goal of 
organisational learning is, therefore, to bring out this humanness (Nonaka, 1998). Humaneness 
Steiro & Evjemo | Learning Embedded in Structures 5
 
 
 
arises in our relationships with others through communities (Plaskoff, 2011). The continuity of 
the community is ensured from a tactical point of view with activities such as meetings, 
distributing information, setting agendas, and facilitating gatherings. The second form is by 
mentoring, namely giving valuable directions (Plaskoff, 2011). Sætren and Laumann (2014) 
found a casual atmosphere and a respectful way of speaking to each other important, for 
example, when someone had or was about to do something not following a procedure. ‘Their 
friendly and respectful interactions resulted in a perception of being taken seriously’ (Sætren 
& Laumann (2014:7). Moreover, Sætren and Laumann (2014) found that the crew relied on 
each other’s collaborative skills and had confidence in each other’s specific task performance. 
Open and respectful communication is imperative when pointing out deviations for 
professionals to be able to learn from this. Lagedec (1993) stresses that good communication 
and organisation need to be established during normal operation.   
Method  
We have conducted an observational and interview study where we have taken inspiration from 
the workplace studies genre as elaborated by Heath, Knoblauch & Luff(2000), which implies 
understanding in depth how naturally occurring ordinary work is performed and organised, and 
the resources used by the participants when accomplishing work tasks. The starting point for 
our analysis is thus to study work practice, which means a focus on the actual accomplishment 
of drilling operations as it is performed in its natural environment on-board the rig. This type 
of study will thus be able to nuance the formalised descriptions of work as they are traditionally 
found in procedures and regulations (Suchman, 1987; 2007). Therefore, the importance of 
understanding the local context related to actual work performance on the rig cannot be 
overestimated. Such an understanding of real versus plan-based work execution is, as 
previously mentioned, acknowledged within the safety discipline as work as imagined versus 
work as done (Hollnagel, 2013). Ose & Steiro (2020) found that the drilling discipline is 
accustomed to planning for contingencies, and it actively participates in meetings and reports 
dangerous elements to their oil company customer. This close collaboration was embedded in 
the processes, and possible outcomes were discussed. 
Phase one (see Table 1) encompassed telephone interviews with ten persons employed on a 
drilling rig with over 15 years of service. The rig can be classified as medium-sized. The rig 
has operated on the Norwegian continental shelf for two to four years. This means that the crew 
had been quite newly employed and that they had recently used the time to familiarise 
themselves with both the technical and social sides of working on this particular drilling rig. 
The crew came from Norway, United Kingdom, Italy and Romania. It is operated by a relatively 
small drilling contractor. Phase two was conducted on a large drilling rig on the Norwegian 
continental shelf, the rig being one of the largest in the world in its class/type. This particular 
rig started operations less than five years ago, and an almost new organisation had to be built. 
The rig is operated by a large drilling contractor. 
A qualitative and explorative approach was applied in our study of offshore drilling. The first 
phase consists of semi-structured interviews with various personnel belonging to a mid-sized 
drilling rig. In phase two, we conducted ethnographically inspired fieldwork on-board a large 
drilling rig during real operations offshore. The interviews gave us insight into drilling 
professionals perceptions and interpretations of actual work practice (Weiss, 1994), while our 
observations provided access to the actual work strategies of the observed professionals—
highlighting in particular informal organisational aspects or the taken-for-granted and 
situational aspects of practice, that is, exploring the perspective of action (Heath et al. 2000).  
 
Steiro & Evjemo | Learning Embedded in Structures 6
 
 
 
Table 1 provides an overview of the two phases of data collection. 
Table 1: Overview of data material 
 Phase one Phase two 
Data collection technique 
 
 
 
What was explored 
Setting 
 
Our informants 
 
Period of data collection 
Data material  
Interviews 
 
 
 
Work and interaction 
Informal setting via phone   
 
Various organisations and roles on 
the rig 
Spring 2014 
Nine hours of data material 
Observational fieldwork including, 
interviews, document studies and 
informal talks. 
Work and interaction,  
formal work setting – naturally 
occurring work 
Various organisations and roles on 
the rig  
Six days in late 2015 
120 hours of data material  
A pilot study (phase one) on one rig provided 10 semi-structured interviews that lasted 45-60 
minutes each. The interviews were fully transcribed and analysed using a thematic approach 
and reported in Steiro, Thevik, & Albrechtsen (2017). Of the 10 informants, eight held 
management positions. The last two were experienced roughnecks, personnel handling manual 
tasks on the drilling floor, making the sample ‘top-heavy’. It was important in phase two to 
work with a broader sample. The current study reports on data material from phase two and is 
based on six days onboard an offshore drilling installation located in the North Sea on the 
Norwegian continental shelf. The two authors carried out ethnographically inspired field 
studies, that is, observational work, semi-structured interviews, and document analysis. The 
pilot study (phase one) was performed by the first author, who also participated in the main 
phase of the project (phase two). The second author had full access to the transcribed interviews 
from phase one.  
The main data collection period (second phase) consisted of a six-day field trip to the offshore 
installation providing both researchers with very rich and context-sensitive data from formal 
interviews, informal talks, and participatory observations —from meetings located inside the 
rig’s administrative quarters to observational periods inside the driller cabin alongside the 
drilling floor. It was not natural, however, nor possible, to use the tape recorder during talks 
with the professionals. Consequently, we always went around together during data collection, 
which meant that we experienced the same events and situations, thus laying the grounds for a 
shared understanding of the empirical material, which was valuable during analysis. We spent 
12-14 hours a day collecting data, making notes, and reflecting on the work observed. This type 
of collaboration enabled the researchers to verify and, in particular, challenge each other when 
collecting and analysing the data on the rig and likewise during further analysis upon returning. 
This was a great advantage in validating the observations and the data retrieved from interviews 
and talks in, for example, the coffee shop on-board the rig, on the drill floor, and the central 
control room (CCR), etc.  
Bryman (2012) claims that both interviewing and taking proper notes can make it difficult to 
catch important information given by the informants. Therefore, being two interviewees 
enabled us to have one to lead and the other to concentrate on taking notes. However, this 
arrangement also allowed the ‘number two’ researcher to intervene and ask questions where 
appropriate, thus following up on themes for clarification purposes or probing deeper into some 
specific theme. Alvesson (2011) writes that one should be open for that meaning is produced in 
and through the interaction between respondent and interviewer. Also, with no tape recordings 
available, we focused on conducting debriefings between us by, for example, sitting down in 
the cabin or coffee shop and going through our notes and collectively reflect on recent events 
and impressions. This is important since Kvale and Brinkman (2009) claim that qualitative data 
Steiro & Evjemo | Learning Embedded in Structures 7
 
 
 
undergo different stages of analysis; during the talks and interviews, after the talks and 
interviews, and finally, when back home, analysing the data and reporting from the findings.  
Thematic analysis was adopted to analyse the interview material. This is a method for 
identifying, analysing, and reporting patterns (themes) within data (Braun & Clarke, 2006). In 
contrast to other types of qualitative analysis (such as grounded theory, interpretative 
phenomenological analysis (IPA), or discourse analysis), thematic analysis is not bound to a 
theoretical or epistemological framework (Braun & Clarke, 2006). Thematic analysis is 
performed through several steps and can be summarised in that the data is taken through the 
process of coding to establish meaningful themes. The actual analysis is not a linear process in 
the sense that the analyst goes backwards and forwards between the data and the codes, as well 
as between the themes and the codes. Our analytical approach was foremost driven by the 
researchers and theoretical interest concerning the research questions and can, therefore, be 
primarily classified as a deductive thematic analysis or a ‘top-down’ way (Braun & Clarke, 
2006). However, the analysis was also explorative, that is, incorporating elements of a bottom-
up analysis of the empirical material, thus also resembling what Alvesson & Skoldberg (2009) 
describe as an abductive analytical approach.  
Results and discussion 
Learning embedded within the PDSA 
We present an overview of the operations related to the PDSA concept. Figure 1 introduces the 
PDSA concept regarding foreseen and unforeseen actions. As we will argue, the organisation 
is continuously working to maintain the plan and avoid contingencies. Simultaneously, several 
of the informants point to both a centralised and formalised structure together with decentralised 
and situated communication as imperative means of operating efficiently and safely. You 
cannot have one without the other.  
 
Figure 1. The PDSA concept seen with both foreseen and unforeseen situations 
From Figure 1, we see the PDSA concept located with the analytical dichotomy of rig work. 
Arguably, the PDSA is an important means to tighten the grip between the foreseen and the 
unforeseen. Drilling operations are very strictly planned; however, what is expected in the 
formation (theoretical assumption) do not always turn out as expected, and adaptions often need 
to be made. Traditionally, adjustments are made primarily as a result of experiences related to 
Steiro & Evjemo | Learning Embedded in Structures 8
 
 
 
‘study’, while operationally at the sharp end, there is empirical evidence for nuance, even 
though these are basically analytical phases. But still, we argue that, based on our data, it is 
natural to emphasise further ‘do‘ through learning as done versus ‘study‘ through learning as 
imagined.  
The informants were very aware of this dualism, and it can be seen as a continuity activity 
(Hernes & Irgens, 2013). Adaptions are not necessarily viewed as interruptions or deviations. 
They can also be viewed as a potential methodological challenge in our study since, according 
to Schein (1985), basic assumptions are best studied during a crisis. However, no such thing 
occurred during our stay, and the drilling installation had operated for a brief period with no 
documented large deviations or accidents. In Table 2, some central situations from our material 
are shown to support the argumentation for Figure 1 and to provide an overview of some 
empirical findings.  
Table 2: Central findings and their explanation 
Situation  Findings Context  Analytical 
1a ‘Are they dancing?’ The driller 
is not fully aware of what is 
going on. He raises his hands 
inside the drilling cabin. Then he 
says, ‘ok’, pushing the stick 
forward and pushing the drilling 
tool further down the well.  
Observation inside the 
drilling cabin. Two 
service companies are 
working on the well on 
the drill floor and the 
drilling crew sitting 
inside the cabin is there 
to act on their 
instructions.  
By stating ‘are they dancing’, 
the driller acknowledges not 
knowing what is going on 
outside the drilling cabin; the 
other personnel present in the 
drilling cabin do not remark or 
comment, and it is our 
impression that none of the 
others knows what currently is 
going on outside on the drill 
floor.  
1b The driller calls the service 
companies to enter the drilling 
cabin. In this ‘time out’, the 
driller tells them upfront that he 
is not familiar with what is going 
on outside and tells them to 
update him on what to do next.  
Observation within the 
drilling cabin. Same 
situation as 1a 
After pushing the tool, without 
fully understanding what is 
going on, he switches on the 
public announcement (PA) 
system and asks the service 
companies to come to the 
drilling cabin. He then asks 
them to give him direct orders. 
The situation is atypical since 
the driller will be fully in 
charge, operating from the 
drilling plan and not taking 
guide/orders from the drilling 
floor.  
1c A good supervisor responsible 
for planning the well, but not 
directly involved in the 
operations enters the drilling 
cabin and monitors the situation. 
He tells us that he has informed 
the two service companies that 
one of them must take the lead in 
the operation. The service 
companies have acknowledged 
this, and one of the persons said 
he would do it. The good 
supervisor turns to us researchers 
and, unsolicited states that ‘this 
is what it is all about all the time: 
planning and communication’.  
Observation in the 
drilling cabin and talks 
with the good 
supervisor.  
Linked to the same 
situation as 1a and 1b 
It is interesting to see situations 
1a and 1b and that 1c is linked 
but independent. The good 
supervisor was not in contact 
with the driller, as we could 
observe before entering the 
drilling cabin. The interesting 
thing is that there are two 
independent approaches to the 
same situation.  
Steiro & Evjemo | Learning Embedded in Structures 9
 
 
 
2 ‘Did it go well?’ Reply made 
calmly in response, ‘It went as 
planned.’  
Observation overheard 
in the locker room. A 
person coming down 
from the offices from a 
small service company 
meets and greets a 
colleague coming in 
from the ‘outdoors’. 
Eagerly ask him about 
the status. The 
colleague replies what 
we interpret as 
confidence in a calm 
tone, ‘It went as 
planned’.   
We see that following plans is 
a way of doing business and 
does not account for great 
excitement when going as 
planned. First, when taking 
notes, we just found it as 
something amusing. However, 
looking back and looking into 
other observations, we find it 
interesting.  
3 This is not directly linked to a 
particular situation but is a 
reflection on different situations. 
The drilling supervisor reflects 
on how, following the drilling 
crew. ‘I am always considering 
the mood and feeling within the 
drilling team.’ When they are in 
good spirit, I try to keep them on 
tip-toe to focus their efforts. 
When in bad spirit, I tell them 
that they are good. I see myself 
as a football manager’.  
Situation 3 is more 
general considerations 
and happened before 
situation 1a-1c.  
In-depth interview with the 
drilling supervisor.  
4 ‘We are constantly trying to 
update the plans to be better 
‘(drilling superintendent). This is 
something they do between 
operations and incorporate in the 
new plans that need to be 
quality-assured. ‘The procedure 
is fresh product; it would have to 
be available here and now. A 
library with procedures does not 
work for me ‘(drilling 
superintendent). 
Situation 4 is more of a 
consideration of the 
role of plans.  
In-depth interview with the 
drilling superintendent. The 
drilling superintendent is 
working onshore and not 
directly following up on the 
drilling operation.  
 
In phase two, we started with an interview with the offshore installation manager (OIM) at the 
rig. He explained the systems they have for learning and knowledge transfer to be a way of 
storing the memory of the organization. This can be seen as a form of procedural memory 
(Huber, 1991; Crossan et al. 1999). He showed us illustrations of the system and pointed to the 
visual management board. It is a board showing numerous indicators of how the rig is 
performing. These results are hanging very visible in the office landscape and close to the main 
meeting room. The same numbers are also reported to the licensee owners. The OIM said that 
a lot was invested in the organisation regarding training and team building. There is a strong 
demand for knowledge transfer reporting, both what works well and what does not work well. 
According to the OIM, the operating company was perceived as quite thorough in the initial 
days when the rig was constructed. At the time, the operator company took a controlling 
approach also into operations. However, the drilling entrepreneur had shown initiative, been 
very open-minded, and as a consequence, trust among the actors increased. The OIM claimed 
that; ‘Now they are confident that we are telling them if there are some problems as early as 
possible. Sætren and Laumann (2014) found trust to be important in high-risk operations during 
Steiro & Evjemo | Learning Embedded in Structures 10
 
 
 
technological change. The OIM further pointed out that the management team had found 
common ground on how to operate. Their work is structured based on the PDSA concept; 
however, they actively adhere to an informal communication style, including quite a lot of 
humour. Of course, this might differ from crew to crew, but there is an informal sense to the 
formality of plan-based work practice. ‘You will see on the meeting that there is no agenda 
presented. But there is a well-defined and known structure. All are familiar with it’. The OIM 
informed us that he participated in approximately 70 meetings during his 14-day offshore 
period. He explained that the drilling entrepreneur had adopted a performance-based 
management structure. Drilling rig operations were to be centred around the concept of Plan-
Do-Study-Act. So PDSA is a framework for planning that also enables communication, where 
communicative ways of conducting collaborative work cannot always be laid down in a 
formalised and standardised procedure. 
The morning meeting between the onshore organisation and offshore installation is the most 
important day-to-day link between them (Haavik, 2011). We witnessed a highly structured but 
also flexible meeting arena with subsequent space and openness to discussions on the rig side. 
Onshore, only the drilling superintendent was actively talking. However, 25-30 persons were 
also present and positioned behind the superintendent. The idea is that they are available if and 
when specific topics occur and to provide information and begin contingency planning if there 
are some reported deviations.  
We also participated in the Sunday morning safety meeting for all personnel not operating the 
rig (the same meeting would be held after dinner and handover for the personnel who had been 
working). Due to cost-cutting, we learned that the meeting was reduced from 60 minutes to 30 
minutes. The camp boss provided soda for all. The atmosphere was positive—a highly 
structured meeting with room for questions and good humour, reinforcing the positive 
atmosphere. There was room for interaction, and management encouraged questions in the 
safety meeting. The meeting ended with the company man explaining the challenges concerning 
well number two. There was reportedly more subsea clay than anticipated. However, both the 
onshore engineers and the crews offshore had managed to drill around the problem. The 
company man explained the integration of knowledge and joint efforts. ‘This operation has 
been a success. We [the operator company] offer a wildlife food evening’. Sætren and Laumann 
(2014) also found a casual atmosphere and a respectful way of speaking to each other as the 
most important factor to facilitate good communication. We found the same in our interviews 
and observation. Plaskoff (2011) points out that an important management task is to provide 
valuable direction in the organisation and see this as an important feature for organisational 
learning.  
The operator company was not so used to drilling entrepreneurs’ involvement in planning. They 
had been sceptical in the beginning, but both informants from the drilling company and the 
operator company stated that this involvement was perceived as very valuable. De Geus (1988) 
observed that scenario planning was not only about planning but that it was equally important 
to get various people to talk actively to each other. Based on our observations, it struck us that 
in our interviews and conversations with the crew about what success meant on the rig, two key 
aspects were mentioned frequently; plans and communication. Several people mentioned these, 
some claiming both aspects as equally important, some put the efforts on plans, others on 
communication. However, we also observed that not all the teams or groups of personnel were 
fully in the loop regarding both plans and communication. The central control room (CCR) was 
placed under the deck quite a distance from rig operations. The two control room operators 
stated that they communicated with everyone within the rig. However, and despite their 
background in drilling, they did not have access to the drilling plan. ‘This makes it difficult to 
assist. A driller can thus expect that CCR is more than what we possibly have the opportunity 
Steiro & Evjemo | Learning Embedded in Structures 11
 
 
 
to be’. The lack of access to the daily drilling program was also confirmed later by another 
control room operator whom we interviewed. Huber (1991) claims that the distribution of 
information influences the degree of organisational learning as it provides more and different 
interpretations of the same information. The drilling program is thus very important but 
seemingly not important enough for the control room to have access to and therefore not in the 
loop. The drilling program’s importance is highlighted by several informants in our study. 
Another example was overheard in the locker room while we were shifting before entering the 
rig’s deck area to visit the driller cabin to observe. An employee from the service company 
‘rushed’ eagerly towards a colleague who just arrived from the drilling deck. Clearly excited 
about what his colleague in orange will convey, he asks with anticipation, ‘Did the job go well?’ 
The colleague responded calmly: ‘It went as planned.’ 
A roughneck from phase one pointed out his contribution regarding safety. ‘The most important 
factor…must be that I report if something might be dangerous. One example is the mud weight. 
I cannot go easy on that, but make sure that I give accurate information on the mud weight’. 
Several of the informants from both phase one and phase two stressed the importance of doing 
things right and following the procedures. It is more important to do things right than fast. Right 
means that one is prepared according to the plan. The right equipment is in place. The driller 
explained that the assistant driller is responsible for preparing tools and equipment for the next 
phase. If this is not done properly, the operation will halt. One toolpusher interviewed in phase 
two reported that procedures together with good cooperation were the most important factors 
for being proactive. Regarding the drilling plan and procedures, the drilling superintendent 
commented that the planning documents should ideally be like this ‘is not too voluminous, 
based on best practice, and communicated properly’.  
Several informants’ states that it was better for all to report if a tool was dropped in the well; it 
was important. This was mentioned by several managers as very important. The rationale could 
be explained by: ‘Then we can handle it and do not get surprised later’ (company man). Jerez-
Gomez et al. (2005) point to four factors regarding the ability to learn: engagement from the 
management, system perspective, openness, and experimentation. In this example, openness is 
the key. Simultaneously, not being punished for a mistake is about clear engagement from the 
management clearly demonstrating no discrepancy between its word and its practice regarding 
the reporting of mistakes (Argyris & Schön, 1978). No fear was reported. However, Haavik 
(2010) reported a lot of trust horizontally but less trust vertically. The latter was attributed to a 
lack of openness in expressing different opinions since it could be a barrier to promotions 
(Haavik, 2010). A basic assumption in continuous improvement initiatives is to create an 
organisational learning environment (Bessant et al., 2001; Wu & Chen, 2006). 
On the operator level, toolbox talks were viewed as the most important tool for safety. It is 
performed for every job and involves all parties. The purpose is to ensure a safe and efficient 
operation by assigning tasks and examining known risks associated with the operation. After 
the action, an after-action review is performed. Regarding procedures, it assesses their 
execution, the resources, and the risk, asking what went well, how to improve, and the lessons 
learned. These are put into a one-pager and becomes part of the PDSA system.  
Drilling work and the importance of following the plan 
Operational morning meetings that involve both onshore and offshore teams are intriguing to 
study since they represent a unique opportunity to gain insight into how the operations are 
planned in advance, communicated, coordinated, and carried out. Specifically, operational 
morning meetings are regarded as the primary arena for collaboration between the onshore rig 
teams, the offshore drilling organisation, and representatives from the service companies 
Steiro & Evjemo | Learning Embedded in Structures 12
 
 
 
onshore (Haavik, 2011; 3–4). The onshore teams possess a variety of resources in the form of 
expertise, software, time to perform calculations, produce documentation, and come up with 
coherent and consistent interpretations with theories, models, and governing documentation for 
drilling. Thus, the onshore teams are specialists in designing and managing plans. On the other 
hand, the offshore crew has their foremost strengths in proximity to operations, their historical 
experience from other wells, and their experience and competence in handling variability at the 
moment. This empirical context thus provides a variety of aspects regarding drilling operations 
(Haavik, 2011), particularly how plans are viewed as imperative for accomplishing successful 
rig operations from the perspective of management. 
However, based on our observations, we also find it instructive to note that the offshore crew 
referred to above manifests itself in two analytically distinct yet interdependent organisational 
entities. Firstly, you have the local management offshore, for example, the oil installation 
manager (OIM) and the drilling section leader (DSL). Secondly, roles related to the drilling 
work include the driller, assistant driller, mud-logger, and roughneck. Although their tasks and 
responsibilities differ, one sees some common characteristics of working within a formal and 
plan-based organisational context. Consider the following examples. 
Formalised, yet loosely coupled meeting arenas, ‘let’s get it right’ 
Let’s look at the operational morning meeting where both onshore and offshore management 
are present. The operating company hosts the morning meeting led by their representative 
onboard the rig, the ‘company man‘. The contractor company rig management participants 
include the oil installation manager (OIM), drilling section leader (DSL), technical section 
leader (TSL), and safety officer. The rig’s assistant manager and HSE coordinator also 
participate from ashore via video. During this meeting, the company man relates primarily to 
his leader ashore, the drilling superintendent, who formally is the highest-ranking person 
present. Along with the operating company are several subcontractors who have placed 
themselves around and behind the drilling superintendent. Some of them are not visible in the 
video feed. The subcontractors are not usually active during meetings, participating only if 
questions related to their areas of responsibility and expertise emerge.  
The morning meeting begins with the company man providing the status of rig operations 
followed by short briefings from some of the contractor employees present. The company man 
continues by asking if anyone has questions, and if there are no questions at the moment, the 
OIM takes over commenting on the focus of the week, that is, ‘toolbox talk‘. According to the 
OIM, daily toolbox talks are very important; ‘In terms of safety, toolbox talk is the most 
important tool‘ (OIM). The company man continues to brief:  
There are no rejected joints so, that’s fine. We also hope for a steady pace now with 
steady joints. It’s time-consuming, but it’s a delicate process. It is going in the right 
direction. Are there any questions about the operations? (Company man). 
The drilling superintendent continues after uttering ‘no‘ in general terms about the progress of 
operations, which is a recap of prior information during the meeting. The OIM takes over once 
more and comments on the plan for today. ‘We will lay down the port forward crane‘(OIM), 
which is endorsed by the drilling engineer and company man. The drilling engineer then 
comments first on the equipment needed for Thursday, referring to the standby vessel in 
particular; ‘There is maintenance on Thursday on xxx. We thus need to have a replacement‘ 
followed by ‘There are no flights today. The crew change flight will be tomorrow‘ (drilling 
engineer). When the morning meeting is about to end, we observe a loose and somewhat 
facetious tone between the parties, initiated by the company man. However, the meeting 
formally ends some moments later with the following comments: 
Steiro & Evjemo | Learning Embedded in Structures 13
 
 
 
Basically, we have all the equipment needed to finalise the well (company man). 
Let's get it right (drilling superintendent). 
Similarly, we observed a somewhat similar distinct meeting structure when we attended the 
work permit meeting. The meeting started at 16.00 hours; seven persons were present in the 
room, including OIM, safety officer, LSD, TSL, and the rig's internal coordinator. The OIM 
initiates by saying, ‘Twenty-one safe talks’ followed by ‘... weekly safety focus, tubular 
handling’. This very brief introduction sets the stage for the subsequent going through of 
observation reports, one example being; ‘... a roll of paper placed on top of the light wall in 
men's toilet on B deck. Action taken: removed paper roll’ (OIM). It strikes us that the work 
permit meeting is structured around the varying persons participating at expected times. 
Nevertheless, it is the OIM who is most active, commenting on what is defined as good 
reporting; ‘Observed good safety culture among roughnecks’. The OIM informs that this has to 
do with roughnecks explicitly asking for permission to enter a zone on the rig. The OIM 
continues with a briefing on the general alarm planned for Sunday morning, followed by a 
discussion on what type of alarm they should initiate. The discussion then continues, focusing 
on the number of people presently involved in upper completion at the rig floor. Thus, safe talks 
are imperative to ensure proper and safe coordination among the involved personnel. 
Furthermore, the safety officer comments that the rig deck seems quite untidy, something that 
needs sorting out. The work permit meeting comes to an end with the OIM commenting on a 
fire alarm earlier that day; an alarm triggered, it seems, by an apprentice flushing water on some 
gas detector outside. This incident causes laughter among those at the meeting.   
Acknowledging situatedness; ‘are they dancing?’ 
Consider the following example (see also Table 2) occurring on the drill deck during good 
completion: Two service companies are currently operating on the drill deck, and we are 
standing inside the driller cabin behind the driller and assistant driller, observing the action. 
The driller and the assistant driller are now responsible for serving and assisting the service 
companies on deck. We are making small talk with both the driller and assistant driller when, 
after a while, the driller, visibly frustrated or annoyed, seems not to know what is happening on 
the drill deck. It turns out that he is unsure of where the service companies are in the drilling 
program resulting in the driller exclaiming in the direction of the assistant driller, ‘Are they 
dancing?’ and throwing out his arms. The driller continues by saying, ‘Yes’ and then pulling 
the joystick forward, controlling the tool supporting the service companies’ work. Shortly later, 
the driller halts. He grabs the microphone and summons everyone on the drill deck inside the 
driller cabin, where he informs them that he has lost track of their progress and that they need 
to inform him right now. The driller is updated, and after a few minutes, the teams operating on 
the drill floor is back up and running. 
Another episode occurred halfway through the conversation with the safety officer. While we 
chatted, an operator from the control informed us via the Public Announcement (PA) system 
that a fire alarm was activated, followed by the operator informing the crew that all work on 
the rig needs to stop while the alarm is investigated. After a few minutes, the PA system calls 
off the alarm, and work resumes. During our talk with the two operators in the rigs' control 
room, they confirmed what revealed during the work permit meeting earlier, that an apprentice 
who inadvertently rinsed water on a sensor caused the fire alarm. The operators told us that for 
all alarms, ‘We always send out someone to check’. 
Furthermore, during our talk with the drilling section leader (DSL), we discussed how important 
it is that employees feel secure about how they perform their jobs. This was something the DSL 
was particularly concerned with in terms of experience and learning. The point is, according to 
Steiro & Evjemo | Learning Embedded in Structures 14
 
 
 
DSL, ‘I must be able to say that this is not good enough, without them out here expecting to get 
fired. That said, people work damn good here’ (DSL). According to the DSL, this varies a lot 
between different countries; however, in this company and on this particular rig, the authority 
gradient is reportedly quite low. DSL continues to reflect on how he prefers to relate to 
subordinates by pointing to what he tends to do concerning pre-job meetings: ‘I'm bored when 
things are routine. Then I go in and try to influence with something new. One is safe as long as 
one is not in a sleep state’ (DSL). He occasionally meets up at these pre-job meetings uttering 
comments like ‘This is not working’, the idea is to basically ‘kick off some thoughts among the 
participants. The DSL tells us that what is most important is to do things the correct way from 
the beginning, meaning that performing work quickly is not considered imperative onboard the 
rig.  
Arguably, key dimensions of successful reliance on plans and procedures (as resources) for 
action in offshore drilling is the trust via transparency among employees, regardless of their 
place in the professional hierarchy. An example of how trust can be facilitated is provided by 
the company man (operator company) linked to losing a utility tool in the well, which will be 
problematic only if the person responsible does not speak up immediately. This transparency is 
extremely important to prevent major problems later. By drawing on the work of Suchman 
(1987) related to plans as resources for action, following plans per se and situated work during 
drilling operations are two sides of the same matters. For example, we have seen above that 
formalised meetings indeed drive the agenda; however, the content of meetings is still situated 
per se. Arguably, the somewhat casual tone within meetings facilitates the loosening of 
couplings in Perrow's (1999) terms. Likewise, actual work out on the sharp end, on the rig deck, 
shows similar characteristics when the driller halts work, calling for an update. An update that 
illustrates situatedness can be a key dimension, even though the drilling plan is the overarching 
management tool. 
Conceptualising the learning dualism—learning as an integrated part of a practice 
Figure 2 is intended to analytically highlight the importance of understanding learning 
processes on the drilling rig as anchored in naturally occurring and hence real work practice—
a work practice that primarily relies on ‘the plan’ as a resource for what is continuously 
performed. At the same time, this work practice is characterised by inherent variation, including 
local adaption, to be meaningful for the professional actors involved. Figure 2 illustrates the 
duality of learning involved when utilising, for example, the drilling plan as a resource for local 
action on the drilling floor. Based on PDSA, including the four stages as a framework for 
organisational learning and, for example, updating the drilling plan—that is, learning per se, we 
argue for the relevance of also recognising conceptually an explicitly situated dimension related 
to learning on the drilling rig. This is the learning that occurs based on the need for local 
adaptions, that is, learning as done, as a parallel to Hollnagel (2014) and work as done. 
Similarly, we argue that learning as imagined reflects how PDSA is currently intended to 
function, which does not cover how learning on the drilling rig manifests itself through actual 
practice. It is worth pointing out that these are two parallel dimensions that are mutually 
dependent on each other, something Hollnagel (2014) also points out through the intention with 
safety I and safety II, that is, respectively, work as imagined and work as done. 
 
 
 
 
 
Steiro & Evjemo | Learning Embedded in Structures 15
 
 
 
Figure 2. PDSA conceptualised concerning the duality of learning 
 
There is an analytical coupling between learning as imagined and expected actions (and 
outcomes) characterised by tight couplings as Perrow (1999) describes the traditional machine 
bureaucracy. Simultaneously, learning as done is grounded in actions characterised by a greater 
degree of uncertainty, that is, unforeseen actions that are traditionally also linked to 
decentralisation (Perrow, 1999) and situatedness (Suchman, 1987, 2007). These two learning 
dimensions are thus to be regarded as interdependent phenomena. Thus, the degree of one or 
the other learning dimension to count for actual organisational learning, as it is analytically 
reflected in the vertical dimension work practices (Figure 2), may vary from situation to 
situation. For applied purposes, it is a matter of the PDSA framework influencing learning 
through practice, at the same time as PDSA is also influenced by the same learning-based work 
practices.    
How to further enable learning in drilling organisations  
As observed and discussed earlier, planning and communication are essential. These can be 
seen as complementary (Bolman & Deal, 1989). Learning is essential for improvement, as seen 
from the theory. A toolpusher provided an example of the importance of a drilling entrepreneur 
supporting the operator at any time. ‘One example is from a casing job in May. The equipment 
that was necessary for the operation was not assembled correctly onshore before shipping it 
offshore. This was discovered when the containers with the equipment arrived offshore. The 
lessons learned were updated in the procedures with a “pre-check” list before sending it 
offshore’. Here we see that these lessons learned as part of the PDSA system resulted in the 
new procedure, thereby embedding the learning into practice. The drilling entrepreneur’s 
operative level gave the input, and the operator used the input to update the procedure for this 
and other company rigs. Imai (1986) points to the importance of spreading the improvements. 
Otherwise, the results will remain local and the industry not optimised (Sobek & Smalley, 
2008). 
‘We are constantly trying to update the plans to be better’ (drilling superintendent). This is 
something they do by and between operations. ‘The procedure is fresh product; it would have 
Steiro & Evjemo | Learning Embedded in Structures 16
 
 
 
to be available here and now. A library with procedures does not work for me’ (drilling 
superintendent). Procedural memory refers to the stored experience, knowledge, and skills that 
are reproduced on a routine basis by organisational members (Cohen, 1991; Moorman and 
Miner, 1998; Singley & Anderson, 1989). In machine bureaucracies like a drilling organisation, 
the drilling superintendent's views are in line with the view of institutionalising knowledge 
(Crossan et al. 1999). A basic assumption in continuous improvement initiatives is to create an 
organisational learning environment (Bessant et al., 2001; Wu & Chen, 2006). One challenge 
is how much information should be put in the drilling program; precise information without 
being too voluminous. Another aspect is the understanding of the drilling program. The drilling 
superintendent points to a solution that works for him: ‘We're talking through what I have 
written. It's about doing putting others` in a favourable position’ (Drilling superintendent).  
After action reviews out on the rig, the engineer writes down the experiences.  ‘We can look it 
over and have it checked out properly’ (drilling superintendent). The drilling superintendent 
meets with the construction manager the day before travelling out to the offshore site for a shift 
period of two weeks (the offshore rotation in Norway is two weeks offshore and four weeks 
onshore). The purpose is to discuss learning and challenges while the drilling manager return 
from his four-week leave. Challenges related to this are all about: ‘To close the loop completely’ 
(drilling superintendent). Get engineers onshore to utilise the learning database and to improve 
plan and procedures. ‘It must be easy for the driller who sits in the chair’ (drilling 
superintendent). ‘We encourage writing a “Tally” (a note or correction in the daily drilling plan) 
and to provide this to the engineer.’ ‘The procedure must function in practice’ (drilling 
superintendent). However, it was claimed from the drillers and driller assistants that what they 
had written not always was taken into the procedures. The drilling superintendent responded 
that this was correct and that there was a delay in time. The onshore organisation needed to 
evaluate and assess risk and have its quality assured. This could take time, but of course, it also 
led to frustration among the operating core. The way forward was to arrange their own meetings, 
collect all the lessons, and feed the newly acquired knowledge into procedures while, at the 
same time, explaining improvement suggestions that had not been taken into account due to 
practical or safety considerations. Hopkins (2000) found that engineers play a crucial role in 
assessing new suggestions for improvement. Some of the frustration reported from the offshore 
side could also be seen not only as a time frustration but also as a lack of feedback and 
explaining why something is not changed due to the request from the personnel operating 
offshore. Husemoen (1997) found that engineering and operations have different perspectives. 
However, these perspectives need to be integrated, although they also can challenge each other 
to find a better-shared solution. Physical proximity was found to be a precondition for success. 
Today, with modern videoconference equipment, physical distance can be reduced. Another 
precondition for success in Husemoen´s study was mutual trust. Trust has also been underlined 
by Sætren & Laumann (2014). Better feedback was needed. A lack of integrations onshore and 
offshore can lead to practical drift (Snook, 2000). Practical drift can be seen as the discrepancy 
between work as planned and work done. A lack of feedback can create frustration with the 
system - blaming the system for procedural practices that obstruct the offshore team´s 
individual insights and experiences are adapted in the procedures (March & Olsen, 1975; Kim, 
1998).  
Haavik (2011) points out that the onshore team has rich resources in the form of expertise and 
software, while the offshore crew has strengths in its proximity to the operations and their 
historical experience from other wells. The PDSA methodology can be seen as following 
similar steps like a scientific methodology (Sobek & Smalley, 2008; Lillrank & Kano,1989). 
That would also imply feedback on suggested improvements.  
 
Steiro & Evjemo | Learning Embedded in Structures 17
 
 
 
Concluding remarks 
We have seen that mutually good communication is essential for both operations and learning. 
Drilling demands thorough planning and rigorous plans. However, changes occur, and 
unforeseen incidents happen. Therefore, not everything goes according to plan, yet the plans 
are still not abandoned per se. However, more interaction and follow-up is needed. We see that 
foreseen actions demand tight couplings. Drilling operations operate within a machine 
bureaucracy well suited for handling and managing tight coupling. However, during unforeseen 
actions, more decentralisation and adaptation of situated action is needed. We see that good 
communication practices must be established during normal operations and extended into more 
demanding phases of the operations (Lagadec, 1993).  
The plan-do-study-act concept (PDSA) used on this rig contributes in both situations. Open and 
lateral communications also contribute, in particular, to problem-solving during unforeseen 
actions. But open and lateral communication during the foreseen actions creates the foundation 
for communication during unforeseen situations with a high degree of complexity. This is done 
not only on the organisation’s top-level but also is performed vertically, i.e., social interaction 
throughout the organisation. Both formal and informal aspects play a crucial role in 
understanding learning as imagined and learning as done. This is important for avoiding 
practical drift (Snook, 2000) and ensuring safe and successful operations. Therefore, the current 
study can be seen as input to reducing practical drift. The key is establishing good practices 
during normal and routine operations, seeing planning as a means of executing and learning, 
communication as means of oiling the machinery and providing feedback into the planning 
processes. The PDSA system in the current study reinforces these processes. 
This paper thus recognises the nuances associated with learning and how offshore learning is 
strongly linked to the execution of planning. It is not new that planning per se is re-evaluated 
and adjusted; it is the way this is done and how this practice-based duality also implies learning, 
which is particularly interesting and relevant to conceptualise. This is because continuous 
variation inherent within real work practice is both a consequence of and a prerequisite for 
learning on-board the drilling rig.  
Acknowledgements 
This work has been performed as part of the project ‘Learning from successful operations’ and 
is sponsored by the Research Council of Norway through the PETROMAKS 2 research 
programme, grant number 228144/E30. No conflict of interest is identified.  
References  
Alvesson, M. & Skoldberg, K. (2009). Reflexive methodology. New vistas for qualitative research. London. Sage.  
Alvesson, M, (2011). Intervjuer: genomförande, tolkning och reflexivitet. Liber. 
Argyris, C. & Schön, D.A. (1978). Organizational Learning: A Theory of Action Perspective. Reading, MA: 
Addison-Wesley.  
Argyris, C. & Schön, D.A. (1996). Organizational Learning II: Theory, Method, and Practice: Theory, Method, 
and Practice. London. FT Press.   
Berger, A. (1997). Continuous improvement and kaizen: standardisation and organisational Designs. Integrated 
Manufacturing Systems, 8, 110-117 doi.org/10.1108/09576069710165792 
Bessant J., Caffyn, S., & Gallagher, M. (2001). An evolutionary model of continuous improvement behavior. 
Technovation, 21, 67-77. 
Bourrier, M. (1998). Elements for Designing a Self- Correcting Organisation: Examples from Nuclear Power 
Plants. In: Hale A & Baram MS (eds.), Safety  
Management. The Challenge of Change. Oxford: Pergamon.  
Brandi, U. & Elkjaer, B. (2011). Organisational learning viewed from a social learning Perspective. Handbook of 
organisational learning and knowledge management, 21- 
41. 
Steiro & Evjemo | Learning Embedded in Structures 18
 
 
 
Braun, V. & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 
77-101. DOI: 10.1191/1478088706qp063oa  
Bryman, A. (2008). Social research methods. Oxford: Oxford University Press. 
Burns, T.E. & Stalker, G.M. (1961). The management of innovation. University of Illinois at Urbana-Champaign's 
Academy for Entrepreneurial Leadership Historical Research Reference in Entrepreneurship. 
Cyert, R.M. & March, J.G. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ. 
Cohen, D. (1991). Individual learning and organisational routine: Emerging connections. Organizational Science 
2(1): 135–139. doi/pdf/10.1287/orsc.2.1.135 
Crossan, M.M., Lane, H.W., & White, R.E., (1999). An Organizational Learning Framework: From Intuition to 
Institution. The Academy of Management Review, 24, 522-537. 
De Geus, A, (1988). Planning as learning. Harvard Business Review 66(2): 70–74. 
Deming, W.E. (2000). Out of the crisis. Cambridge, Mass., MIT Press. 
Goh, S. & Richards, G. (1997). Benchmarking the learning capability of organisations. European Management 
Journal, 15, 575-583. doi:10.1016/S0263-2373(97)00036-4 
Heath, C., Knoblauch, H. & Luff, P. (2000). Technology and social interaction: the emergence of workplace 
studies. British Journal of Sociology, 51(2): 299-320. 
Hellesøy, O.H. (Ed.). (1985). Work Environment, Statfjord Field: Work Environment, Health, and Safety on a 
North Sea Oil Platform. Oslo: Universitetsforlaget. 
Hernes, T. & Irgens, E.J. (2012). Keeping things mindfully on track: Organisational learning under continuity. 
Management Learning, doi: 10.1177/1350507612445258. 
Hollnagel, E. (2014). Safety-I and Safety-II. Farnham: Ashgate. 
Hollnagel, E. (Ed.). (2013). Resilience engineering in practice: A guidebook. Ashgate Publishing, Ltd. 
Hollnagel, E. (2009). The ETTO principle: efficiency-thoroughness trade-off: why things that go right sometimes 
go wrong. London: Ashgate Publishing, Ltd. 
Hopkins, A.H. (2000). Lessons from Longford: The Esso gas plant explosion. Sydney: CCH Australia Limited. 
Huber, G.P. (1991). Organizational Learning: The Contributing Processes and the Literatures. Organisation 
Science, 2, 88-115. 
Husemoen, M.S. (1997). Bridging operation and design. The encounter between practical and discipline-based 
knowledge in offshore platform design. 
Haavik, T.K. (2011). Chasing shared understanding in drilling operations. Cognition, Technology & Work. 13(4), 
281-294. DOI: 10.1007/s10111-010-0166-z 
Haavik, T.K. (2010), Making drilling operations visible: the role of articulation work for organisational 
safety. Cognition, Technology & Work, 12(4), 285-295. Doi:10.1007/s10111-010-0139-2.  
Imai, M. (1986). Kaizen (Ky'zen): the key to Japan's competitive success. Random House. 
Jerez-Gomez, P., Céspedes-Lorente, J. & Valle-Cabrera, R. (2005). Organisational learning capability: a proposal 
of measurement. Journal of Business Research, 58, 715-725. http://dx.doi.org/10.1016/j.jbusres.2003.11 
.002 
Kim, D.H. (1998). The link between individual and organisational learning. The Strategic Management of 
Intellectual Capital, 41-62. 
Kvale, S. & Brinkmann, S. (2009). Interviews: Learning the craft of qualitative research Interviewing. London: 
Sage.  
Lagadec, P. (1993). Preventing chaos in a crisis. Maidenhead: McGraw-Hill. 
LaPorte, T.R., & Consolini, P.M. (1991). Working in practice but not in theory: theoretical challenges of ‘high-
reliability organisations. Journal of Public Administration Research and Theory, 1(1), 19-48. 
March, J. G. (1991). Exploration and Exploitation in Organizational Learning. Organisation Science, 2(1), 71-87. 
Doi:10.1111/ijmr.12076.  
March, J.G., & Olsen J.P. (1975). The uncertainty of the past: organisational learning under Ambiguity. European 
Journal of Political Research, 3, 147-171. 
Mintzberg, H. (1979). Structures in fives. Prentice-Hall. 
Moorman, C. & Miner, A.S. (1998). The convergence of planning and execution: Improvisation in new product 
development. Journal of Marketing 62(3): 1–20. DOI: 10.2307/1251740. 
Morgan, G. (1980). Paradigms, metaphors, and puzzle solving in organisation theory. Administrative science 
quarterly, 605-622.  
Morgan, G. (2006). Images of Organization. Sage. 
Morgan J.M. & Liker, J.K. (2006). The Toyota product development system. The Productivity Press.  
Nonaka, I. & Konno, N. (1998). The concept of ‘ba’: Building a foundation for knowledge creation. California 
management review, 40(3), 40-54. 
Nonaka, I. (1994). A Dynamic Theory of Organizational Knowledge Creation. Organisation Science, 5, 14-37. 
DOI: 10.1287/orsc.5.1.14 
Steiro & Evjemo | Learning Embedded in Structures 19
 
 
 
Ose, G. O., & Steiro, T. J. (2020). An Analytical Framework for Resilience Exemplified With a Real-Time 
Operational Center. ASCE-ASME J Risk and Uncert in Engrg Sys Part B Mech Engrg, 6(1). 
Lillrank, P.M. & Kanō. N. (1989). Continuous improvement: quality control circles in Japanese industry. Center 
for Japanese Studies University of Michigan. 
Pedler, M., Burgoyne, J. & Boydell, T. (1999). The learning company: A strategy for sustainable development 
(3rd ed.). McGraw-Hill. 
Pedler, M., Burgoyne, J. & Boydell, T. (Eds.) (1990). Self-development in organisations. McGraw-Hill. 
Perrow, C. (1984). Normal Accidents: Living with High-Risk Technologies. Basic Books.  
Plaskoff, J. (2011). Intersubjectivity and community-building. In: M. Easterby-Smith, & M. A. Lyles (Eds.), 
Handbook of organisational learning and knowledge management (2nd ed., pp. 199–223). Wiley. 
Rasmussen, J. & Svedung, I. (2000). Proactive Risk Management in a Dynamic Society. Swedish Rescue Services 
Agency.  
Rosness, R., Grøtan T.O., Guttormsen G., Herrera, I.A., Steiro, T., Størseth, F., Tinmannsvik, R.K. & Wærø, I. 
(2010). Organisational Accidents and Resilient Organisations: Six Perspectives. SINTEF.  
Shook, J. (2008). Managing to learn: Using the A3 Management Process to Solve Problems, Gain Agreement, 
Mentor and Lead. Cambridge. MA: Lean Enterprise Institute.  
Simon, H. (1991). Bounded rationality and organisational learning. Organization Science, 2(1), 125-134.  
Singley, M.K. & Anderson, J.R. (1989). The Transfer of Cognitive Skill. Harvard University Press. 
Snook, S.A. (2002). Friendly fire: The accidental shootdown of US Black Hawks over Northern Iraq. Princeton 
University Press. 
Sobek, D.K. & Smalley, A. (2008). Understanding A3 thinking: a critical component of Toyota's PDCA 
management system. CRC Press. 
Steiro, T. J., Thevik, A., & Albrechtsen, E. (2017). On Formal and Informal Factors: Enabling Learning for Safe 
Offshore-Drilling Operations. International Journal of Management, Knowledge and Learning, 6(2), 85-
106. 
Suchman, L. (2007). Human-machine reconfigurations: Plans and situated actions. Cambridge University Press. 
Suchman, L.A. (1987). Plans and situated actions: the problem of human-machine communication. Cambridge: 
Cambridge University Press. 
Sætren, G.B. & Laumann, K. (2015). Effects of trust in high-risk organisations during technological 
changes. Cognition, Technology & Work, 17(1), 131-144. DOI 10.1007/s10111-014-0313-z 
Von Krogh, G., Ichijo, K. & Nonaka, I. (2000). Enabling knowledge creation: How to unlock the mystery of tacit 
knowledge and release the power of innovation. Oxford University Press. 
Weiss, R.S. (1994). Learning From Strangers: The art and method of qualitative interview studies. Free Press. 
Womack, J.P. (1990). The machine that changed the world. Simon and Schuster. 
Wu, C.W. & Chen, C.L. (2006). An integrated structural model toward successful continuous improvement 
activity. Technovation, 26, 697-707. http://dx.doi.org/10.1016/j.technovation.2005.05.002 
 
Steiro & Evjemo | Learning Embedded in Structures 20