Introduction In safety and other aspects of organizational life, often the only time we actively learn about what s going on is when something bad happens. For example, after an accident we actively look back to figure out what happened. We do this so we can fix whatever caused the accident, so that the accident never happens again. The rest of organizational life is spent doing. We assume that everything is working according to plan, so there is no need to spend time learning about what is happening we determined that when we created the plan. In effect, no news is good news. But are we tying one hand behind our backs when we take such a stance? In this workshop we will discuss this belief and present a different perspective, one that sees failure as merely the other side of the coin from success. And if that s true then not only is there more opportunity to learn about what s going in our organizations, it is important that we do so. Let s get started! Appreciative Investigations 1 SCM 2017
Understanding Failure, Success, Normal Work, & Appreciative Investigations A common way of explaining accidents, or other failures in organizations, involves the concept of root cause the idea that if you dig deep enough, you can find the primary or fundamental cause of an accident. Different people have different theories about what a root cause is, but all theories of root cause have one thing in common if it wasn t for the root cause, the accident would not have happened. Basically, everything in the organization was working fine until the root cause came along. Typically the root cause is seen as a failure, either in the individual directly involved in the accident, or in the actions of management, or the management system. This idea that something has to break or fail for an accident to occur is based on a fundamental belief in the bimodal functioning of systems, illustrated in the figure below. Whenever things function normally, as we had planned, then successful or acceptable outcomes result. However, when things begin to malfunction (i.e., when something breaks or people start breaking rules and deviate from our plan) that s when failure happens and we get unacceptable outcomes, like accidents. Normal functioning (Everything works) Success/Acceptable Outcome! Unwanted transition Malfunctioning (Something breaks) Failure/Unacceptable Outcome " (Hollnagel, 2012) In this model, you can see that there s no need to do any investigation or learning when things are successful, because everything is happening according to the plan. When we have a failure though, such as an accident, then we need to investigate what went wrong so we can find the thing that broke and fix it. In essence, the bimodal functioning model sees accidents as having completely different underlying causes and conditions than normal, successful function. Put simply accidents and successes have entirely different causes. Appreciative Investigations 2 SCM 2017
Appreciative Investigations But is this true? The short answer is yes, but only if we re talking about machines. Think of a clock that breaks so that it no longer tells time accurately. Why does that happen? Because some parts within the clock stopped working the way they were supposed to. If everything were working then the clock would work. But once certain parts stop working to a certain level the clock stops working. That s bimodal functioning. Success and failure have different underlying causes or conditions. However, this model doesn t work when you start to involve people. People don t People don t fail like machines. To illustrate, lets use the example of someone who is walking down the street and collides with a fail like lamppost because they didn t see it, as their mind was on something machines. else. Often accident investigations will note that the cause was inattention, but this is misleading. Conscious human beings are always paying attention to something. The person who collided with the lamppost wasn t paying attention to what was in front of them, but they were paying attention to whatever was on their mind (their job, their family, the game, etc.). Paying attention to something else while you re engaging in an activity is a normal thing that people do everyday. In fact, if you polled the other people on the street who didn t run into the lamppost you d likely find others who also weren t paying attention, but who did not run into the lamppost. What are we to make of this? Well, this means that the same conditions that were present during a failure (the person who ran into the lamppost) were also present during non-failure, or success (the people who didn t run into the lamppost but who were also not paying attention). This puts serious doubt into whether the bimodal functioning modal works for people. To add to this doubt, we should ask why the people were not paying attention to the lamppost? The answer is that people often have to engage in mundane activities that don t require much thought, so to increase efficiency we often engage our minds in thinking about something else. This multi-tasking is something we do to make ourselves more effective. Not only does this make us less bored, but it also increases our overall performance by saving time in thinking about issues later (not to mention that thinking intensely about some tasks actually increases the number of errors we make, e.g., typing, so thinking about something else may actually decrease errors overall). Again, what does this mean? What we see emerging is that the things that make people fail not only are present in times of success, but also increase the probability of success. As Professor Erik Hollnagel notes, people adjust their performance to their environment in order to increase the probability of success and decrease the probability of failure. As shown in the figure below, these performance adjustments work most of the time, but occasionally also lead to failure. Appreciative Investigations 3 SCM 2017
Performance Adjustments causes Success/Acceptable Outcome! In essence, the things that cause success with people are similar or the same as those things that cause failure. Put simply success and failure are two sides of the same coin. Appreciative Investigations Hollnagel (2012) Failure/Unacceptable Outcome " This has profound implications for how safety and organizational professionals learn. It means that we should be able to learn just as much from successful outcomes as we do from failure. So why wait for an accident to happen to start the learning process? No news is not good news. The same goal conflicts, inadequate resources, improper tools, poor training, and unrealistic procedures we find after the accident are present before the accident. Let s go out and find them! This process of learning from normal work processes is known as an Appreciative Investigation, sometimes called Positive Investigations or Success Investigations. Appreciative Investigations are different than other safety tools, such as behavior observations or audits. Typical safety tools are merely looking to see if the organization and its people are following an agreed upon standard (e.g., a regulatory standard or a procedure). Observations and audits are not really interested in how work is done, but in whether the work is done in accordance with the standard. Appreciative Investigations, by contrast, are interested not so much in what should be, but in what is. They focus on the performance adjustments workers have to make to get the work done in an imperfect and complex organization. Once these adjustments are identified the organization can then evaluate them to see how to improve the processes. Appreciative Investigations can also identify hidden innovations that workers have created to overcome problems, which can be shared throughout the organization once identified. One key benefit of Appreciative Investigations is that they not only allow the organization to identify issues that are traditionally considered safety issues (e.g., unknown hazards and risks), but by focusing on how the organization can better support and set workers up for success, the Appreciative Investigation aligns the goals of safety with other organizational goals, such as production, quality, and environmental. There are four basic types of Appreciative Investigations. Each of these are discussed respectively in the next four sections. Appreciative Investigations 4 SCM 2017
Debriefs Basic Description Depth Time in Relation to the Task A semi-structured discussion with workers following the completion of a task or project. Low to Medium Immediately following the task Advantages Not very intrusive Easy to facilitate Disadvantages Limited by the workers openness Generally time constrained Typically doesn t allow for in-depth analysis Recommended Methodology 1. Dedicate about ten minutes (up to an hour for more complicated projects) at the end of a work shift or project for one work crew or project team. 2. Gather the group in a room, ideally with a dry erase board, chart paper, or the ability to post sticky notes for note taking purposes. Identify someone to take notes during the session. 3. Ask at least these three questions to the group: What worked or went well? What didn t work or didn t go well? What surprised you? 4. Ask the questions one at a time, allowing sufficient time between each question for the group members to answer. 5. Ask follow-up questions to clarify information. Try to draw information out of all group members if possible, without being pushy. 6. At the end of the session, summarize any interesting lessons learned and identify any necessary follow-up actions. Ensure that each follow-up actions are provided to appropriate responsible person(s). Appreciative Investigations 5 SCM 2017
Post-Task Investigation Basic Description Depth A semi-structured group dedicated to conducting an in-depth learning about the task or project. Medium to High Time in Relation Hours or Days (sometimes longer) after completion of the task or project to the Task Advantages Provides for significant amounts of reflection and sensemaking for all involved Builds trust and engagement in those involved Process lends itself to identify significant improvements and lessonslearned Some flexibility in how the analysis is conducted Disadvantages Requires time from multiple people Requires experienced facilitator to guide the process Sometimes subject to faulty memory Recommended Methodology 1. Identify a project or task that would benefit from some in-depth analysis. 2. Identify who should be on the team. You definitely want representatives of those who do the work, as well as some diverse perspectives (supervisors, engineers, etc.). 3. Identify a good location to house the team for a few hours, ideally a place that is private but near where the work is conducted to allow for field trips. You also want a space that has a dry erase board, chart paper, or space to put up stickies for note taking. 4. Start the session by explaining the goals and the process. Identify who is going to take notes during the session. 5. Start early in the work process, ideally where the work begins, typically at the beginning of the day, and move forward in time. 6. Ask participants to share their unique perspectives on how the work is performed. Sometimes even asking very broad questions (e.g., so, what is your role here?) and narrowing down your questions as the discussion moves forward helps to warm people up. 7. Ask clarifying and follow-up questions to draw information out of participants about what and how they do the work. Use the questions primer as needed. 8. Do not allow anyone to start solving problems yet. Just focus on understanding the context around the work that is done. Appreciative Investigations 6 SCM 2017
9. Once the team feels that they have a good understanding of the project or task that was performed (the goals, the conflicts, the resources necessary, etc.), move into the next phase. 10. If possible, have soak time of one hour to one day. Make sure you discuss with all participants when and where to reconvene before going on soak time! 11. Once you come back, review the story of the task or project and what you all learned. Allow people to inject new information or ask new questions. Do not move to identifying improvements until all new information is explored. 12. Once the previous phase is done, ask the group how the task or process can be improved in the future so that the people involved are set up better for success in the future. How can we make the work easier? How can we ensure that all tools and resources are readily available? How can we ensure that people involved have the competence they need? 13. Capture any improvements or innovations identified in the process and, as a team, assign to those responsible within the team to ensure that these get implemented. Assist the team members in this process, but do not take over the improvement process. Allow them to own it. Appreciative Investigations 7 SCM 2017
Task Observation Basic Description Depth Time in Relation to the Task Analysis of work being conducted in action using semi-structure or unstructured methods (can include follow-up interviews) Low to High (Depends heavily on the observer and the methods used) Conducted during work operations Advantages Does not merely rely on worker memory or knowledge Can lead to some in-depth conclusions when done by a knowledgeable observer Usually not very intrusive Disadvantages Requires significant trust between observer and workers Observer needs to have access to domain-specific knowledge (can be gained through interviews) The simple presence of the observer may change the work being observed Recommended Methodology 1. Identify a task or project you d like to observe (this can also be done opportunistically, during a site walk). 2. Ensure that your observation can be conducted at a safe and nonobtrusive distance. 3. Introduce yourself to those you will be observing and explain the purpose of your observation and what they can expect. Emphasize that you are not there to judge their work, but to help improve the work processes. 4. Take notes and pictures of the work being done, if possible (obtain permission for all pictures taken). 5. Note the work steps you observe. When possible (either during the observation or in reflection afterwards) analyze the work step to identify what the goal of each step is and what this implies about the work processes, constraints, resources available, etc. 6. If possible, set aside a long period of time to observe the work, ideally hours. 7. Upon completion of your observation, obtain an interview with the observed workers to describe what your observations. Allow them to react to each observation and provide additional context and analysis, where appropriate. Use the question primer as needed to identify appropriate questions to ask. Appreciative Investigations 8 SCM 2017
8. If appropriate and necessary, take your observations and analysis and share them with other workers who do the similar task to get their perspective on what you saw. This will help validate your observations. 9. Develop a set of recommendations designed to better set the workers up for success whenever possible. Discuss these recommendations with the observed workers and gain their approval before submitting them for implementation. Appreciative Investigations 9 SCM 2017
Job Shadowing/ Day in the life of Basic Description Depth Time in Relation to the Task Following a worker or group of workers for part of or an entire shift and conducting the work with them whenever possible. Medium to High Conducted during work operations Advantages Allows for an embedded view of the work processes The participant-observer gets to experience the work as similarly as possible to how the workers experience it Often builds significant engagement and relationships with workers Disadvantages May require the participant-observer to do dangerous work or work that he/she is not qualified for Can be intrusive, as the workers have to watch out for the participant-observer May not provide an accurate view of how the work is performed if the workers adjust their work to account for the participant-observer Recommended Methodology 1. Identify the task or project you d like to observer with sufficient advance notice to allow you to ensure that you have the required training, equipment, and planning to ensure a thorough analysis. 2. Introduce yourself to the workers who you will be observing. Provide sufficient time to explain the process and allow them to ask any questions. Make sure it is clear that you are working under their direction, regardless of your normal position in the organization. 3. Conduct the observation. Take notes and pictures (with permission), whenever possible of work processes or issues that you d like to analyze later. Ask questions (using the question primer, as needed) during the time to dig deeper into the work processes. 4. At the end of the time, debrief with the workers to share yours and their observations. Collaboratively identify opportunities for improvement and lessons-learned and ensure that these items are followed-up on. Appreciative Investigations 10 SCM 2017
Question Primer Appreciative Investigations The question is the investigator s most important tool. The questions we ask dictate the answers we get, which in turn dictates the conclusions we draw. Below are some questions you can ask to draw information out of people and dig deeper into work processes. These are merely suggestions. You do not have to ask all of the below questions. Just use what you think would be useful. And feel free to create your own questions. Cues Interpretation Errors Previous Experience or Knowledge Goals Taking action Outcome General What were you seeing? What were you focusing on? What were you expecting to happen? If you had to describe the situation to your coworker at that point, what would you have said? What mistakes were likely at this point? Were you reminded of any previous experience? Did this situation fit a standard scenario? Were you trained to deal with this situation? Were there any rules or procedures that applied clearly here Did any other sources of knowledge suggest what to do? What were you trying to achieve? Were there multiple goals at the same time? Were there time pressure or other limitations on what you could do? How did you think you could influence the course of events? Did you discuss or mentally imagine a number of options or did you know straight away what to do? Did the outcome fit your expectation? Did you have to update your assessment of the situation at any point? What is something that makes this task difficult? What is something about this job that we do that just doesn t make sense? Do we have all the resources we need in order to do this job safely? What s something we could do to make this task safer or easier to do? Are there some things about this task that are unduly risky or difficult that we just put up with or have gotten used to? If you could improve one thing about doing this task, what would it be? Do we have procedures or rules related to this task? If yes, are they realistic and easy to follow? Do you ever have to find a way around the procedures and/or rules to get the task done? Appreciative Investigations 11 SCM 2017
Is there training associated with this task? If yes, how closely did the training match the reality of completing the task? If someone were to be seriously hurt doing this task, what would the cause be? What s something that you wish management would do that would make this task safer or easier to complete? What would you recommend? Appreciative Investigations 12 SCM 2017
Suggested Reading Appreciative Investigations 13 SCM 2017