Research project aims to improve safety by learning from success

Maersk/SINTEF collaboration to inform industry on actions, patterns that contribute to safe, successful operations

By Ranveig Kviseth Tinmannsvik and Stian Antonsen, SINTEF

Many theories, concepts and perspectives are available to the petroleum industry to help it understand failure. Similarly, there is a rich array of methods to analyze the causes of accidents and critical events. A great effort is put into such analyses and dissemination of results, by the industry and by national authorities. However, when it comes to successful operations, the industry lacks the language and theoretical “infrastructure” to conceptualize success. Further, there is a scarcity of methods for analyzing and learning from successful operations.

To correct some of this shortcoming, SINTEF and Maersk Drilling are collaborating on a research project on learning from successful operations, funded by the Research Council of Norway.

Defining success

It may seem straightforward to think of success as the opposite of failure. However, the notion of success may be ambiguous. Several different criteria have been used to identify an operation as successful with regard to safety in research literature:

1. Absence of adverse consequences. An operation may be labeled successful if it did not directly lead to an accident.

2. Lower frequency of adverse consequences than other similar operations under similar conditions. Success implies that an operation is performed with fewer instances of adverse consequences than other units performing the same operation under similar conditions (a benchmarking perspective).

3. Lower frequency of adverse consequences than would be expected, taking into account the nature of the operations. This seems to be the understanding of success in early studies of high-reliability organizations (HROs).

4. Low frequency of adverse consequences in an absolute sense or compared with a broader range of operations. This can be exemplified by studies of passenger fatalities in transportation. One transportation sector can be characterized as “ultra-safe,” based on a very low number of passengers killed per person-kilometer, compared with other means of transportation.

5. Successful coping close to the boundary of safe operations. An operation is labeled successful if the actors manage to operate with very small safety margins without experiencing an accident.

6. Successful recovery from an imminent danger. An operation may be labeled successful if the actors managed to recover from a potentially disastrous sequence of events. One may prefer to speak of “successful recovery” rather than “successful operations” in this case.

7. Maintenance of a broad safety margin. An operation may be labeled successful to the extent that a broad safety margin was maintained throughout the operation.

8. Maintenance of an effective barrier structure. An operation may be labeled successful if an adequate set of barriers against adverse event sequences was in place throughout the operation.

9. Maintenance or improving a high level of safety over a long period of time. This implies resisting the dangers of “drift into failure,” where safety margins are eroded over time.

Most of the criteria used in the research literature are related to outcomes, i.e., the absence of or low frequency of adverse events. However, criteria 7 and 8 are related to processes that are assumed to be closely related to the potential for adverse outcomes. A common pitfall when characterizing an operation as successful is to generalize from one category of accidents to another category that is only weakly correlated. For instance, successful prevention of lost-time incidents does not necessarily imply that the risk of major accidents is under control, as illustrated by the Macondo disaster.

Challenges in learning from successful operations

Learning from success has much to do with understanding variability. Basically, we assume that there will always be some variation in the way work is carried out and decisions are made. Therefore, it becomes important to understand how operators, policymakers and technical systems are able to handle this variability in a safe manner.

A simplified model for success, failure and variability – as well as how we handle deviations from normal variation in work performance – is shown in Figure 1. In a specific work operation, we operate within this room of maneuver. Hopefully, most of the time, the operation will operate in the central field, with substantial safety margins (green zone).

Occasionally, the operation will approach the critical area (red zone). Therefore, we need to have expertise and a contingency preparedness to ensure a successful recovery from failure. These corrections are marked with a “barrier” symbol. In addition, we have minor adjustments that happen in everyday situations to ensure successful operations.

Safety and successful operations is about the absence of problems and can be perceived as something invisible (a non-event). When nothing happens, there is no need for action and there is no sense of urgency. Another challenge is that it can be difficult to explain a success in such a way that we can learn from it.

How do we get people to reflect on success when “nothing” happens? An important premise for learning from success is to foster an understanding that when “nothing” happens, a lot of things are actually happening that prevented things from going wrong. Our experience is that, in addition to interviewing people, we need to observe their collaboration and sensemaking processes in the execution of work. This is because the operating personnel perform many actions to control safety without recognizing it as a successful operation. For them, it is more seen as a normal way of working.

Case studies

The empirical basis for the joint project is offshore drilling operations. Drilling operations differ from other offshore operations by its dynamic nature, a high level of uncertainty and the large number of involved actors collaborating in the work. Offshore drilling is often referred to as a continuous process of problem-solving, where new and unexpected situations arise and must be managed on the spot. Therefore, drilling operations represent a good basis for studying learning from successful operations.

A qualitative and explorative approach is applied in our study of offshore drilling:

1. Observations of simulator training;

2. Offshore observations (meetings, work, interactions, routines); and

3. Interviews offshore/onshore.

Observation of simulator training took place at Maersk Training’s drilling simulator in Svendborg, Denmark, involving observations of actions, interactions and practices promoting safety. The interviews and observations onshore/offshore were conducted in two phases. Phase 1 was a pilot study entailing onshore semi-structured interviews with drilling crew personnel operating a medium-sized drilling rig. Phase 2 was a six-day field trip onboard a large offshore drilling facility located in the North Sea, providing the researchers with rich data from interviews, informal talks and participatory observations from meetings and in operations.

What characterizes a successful operation?

Planning and communication

An important prerequisite for successful operations is to be aware of the safety margins and know where you are relative to the limits of safe behavior and decisions. The informants describe a successful operation as an operation that is well planned, where small problems are solved continuously as they come up, where consequences of actions are examined in advance, where all involved personnel are well prepared to cope with unexpected situations and with well-functioning rules and procedures.

An informant’s view on how he works to avoid major accidents:

“It must be that I always inform about things that I see, or think might be dangerous with both the equipment and staff. And when in operation, I do not look lightly on things like mud weight. I constantly give correct information so that those who govern it have the right information” (roughneck).

More of the informants emphasized the interplay between planning and the need for communication. The drilling superintendent claimed that the road to success is characterized by “plan plus knowledge exchange, then it will usually go well.” The companyman emphasized: “Always look forward in time.”

Dealing with uncertainty and ambiguity

Successful operations is about handling deviations from expectations and dealing with uncertainty and ambiguity in a sensemaking process. Uncertainty occurs when we are not able to make sense of a situation, while ambiguity means that different interpretations are possible. Important elements of a sensemaking process are shown in Figure 2. When something deviates from expectations, we have two choices. Either we try to explain away the anomaly as a coincidence (reference the “bladder effect” in the Macondo disaster), or we use it as an opportunity to check the validity of our expectations.

Successful operations imply being wary of the dangers of expectations and remaining vigilant to anomalies and treating them as critical indicators of potential and emergent problems. We use the term “anomalizing” to mean taking proactive steps to become alert to discrepancies, to understand them more completely and to be less encumbered by history. Knowing when to act and how to act on a weak signal can make the difference between success and catastrophe.

The nature of interpreting is associated with social processes to cope with uncertainty: A process where different bits and pieces of information are combined to produce an image of what is going on. The next step is to use the information from the interpreting to anticipate possible future development scenarios. This is essential to be able to see things coming and to be prepared to cope with the situation that evolves.

Sensemaking, e.g., in a rig team, is a social process that is triggered by uncertainty or ambiguity. It is an ongoing process without a beginning or end. Paying attention to uncertainty and ambiguity by making sense both of the past and the future will be an important precondition for success.

Methods and guidelines

A main deliverable from the project is a debriefing/observation guide to help organizations learn from their successful operations. The objective of the guide is to help practitioners and researchers identify contributions to safe operations in order to learn from operations that are performed without adverse consequences, with adequate safety margins and with appropriate barriers in place. A first version of a debriefing guide has been developed.

The approach of the guide is to help sensitize the user to actions and patterns of interaction that may contribute to successful operations. We focus on how actions or interaction patterns may contribute to successful operations. Our objective is to help the observer “see more,” i.e., notice and interpret episodes that might have gone unnoticed without this guide.

The guide will be completed by summer 2016 and will be communicated to practitioners through a workshop. The publication will be made available for downloading from SINTEF’s website. DC

This article is based on a presentation at the 2015 IADC Drilling HSE&T Europe Conference, 23-24 September, Amsterdam.