Safety-II, its cousin Resilience Engineering (and offshoots such as resilient healthcare), as well as predecessor concepts and theories, have attracted great interest among organisations and their staff. People, especially front-line staff, understand the need to understand all outcomes – wanted and unwanted – and the systems and associated patterns of system behaviour that generate these outcomes. The trouble is, people are not sure where to start with ‘doing Safety-II’. Some methods and seemingly complicated words and ideas might seem off-putting. They don’t need to be. In this post, I will provide some initial ideas and inspiration for getting started. The ideas are in plain language without reference to any specific techniques.
Idea 1: Collaborate
Safety-II and Resilience Engineering are not solo efforts. You can do little of practical benefit alone. In fact, going alone will almost guarantee a miserable work life. You will start to see the reality of how patterns, system structures and mental models are connected to produce events, both wanted and unwanted. But you will have to stand back and watch how this complexity is boiled down to mechanistic thinking and methods that don’t describe how safety is created, or even how unsafe events really occur. You will also have to observe foes of intervention in action, which almost guarantee unintended consequences. For the sake of sanity, it is almost better not to know how complex systems fail, let alone how they work on a day-to-day basis. Finding a small number of open-minded people who are willing to expand their thinking and listen to ideas and experiences without prejudgement, and not hamstrung by personal barriers, is a good place to start. A diverse group that traverses organisational silos is helpful.
Idea 2: Read
If you want to do Safety-II, you have to read. At least a bit. You might find that you don’t have enough time to read technical books. You don’t have to, though you may well want to, at some point. Start by reading some short articles on Safety-II, and associated concepts, by authors with a pedigree in this area. You might want to expand your search terms to ‘systems thinking‘, ‘resilience engineering‘, ‘systems ergonomics and human factors‘. From here you might start to explore methods from social science (e.g., action research, practice theory, ethnography). See where the search takes you, from blog posts (search this blog for a few, as a start), through to White Papers, articles (email the author if you can’t access them), and books. A couple of short articles a week and you’ll be on your way to understanding the key ideas. Be mindful that some of what is written may be way off the mark (what Safety-II isn’t), as Safety-II, like anything else, is subject to the bandwagon effect.
Idea 3: Think
It might seen strange to suggest thinking as a way to do Safety-II or Resilience Engineering. But in many lines of work, we somehow manage to avoid taking a step back to think more holistically about outcomes, work, systems, and the mental models that give rise to all of this. I teach a systems thinking course which is about…thinking. At the end of the most recent course, one participant said that it was the first course that they had participated in where they actually had to think, and not just learn content or follow a process. The course doesn’t provide a process, but rather a space to think and challenge one’s own assumptions. The thinking required involves going up and out to the system as a whole, switching perspectives (stakeholders and situations), and generally questioning how things go. Thinking through situated examples is especially useful, so long as there are links to theory.
Idea 4: Listen and Talk
From the above, and the below, prepare some topics or questions on concepts, methods and everyday work for discussion. Find a room, get some drinks and snacks, and arrange some chairs in a circle. Try to get rid of tables and anything else that gets between you. The questions may emerge from your reading or from your experience…preferably both. E.g. If you had to explain to a neighbour why your organisation operated safely, what would you say? What do we do well? What dilemmas do we face? What surprises do we experience? How do we handle them? What unintended consequences have we experienced from interventions? What factors are at play when things go right and wrong? What is the role of designed artefacts and processes versus adaptive performance in creating safety? A good discussion will harvest new insights, including multiple perspectives and thick descriptions.
Idea 5: Write and Draw
Write about your experiences of work in the frame of Safety-II or Resilience Engineering. Think deeply about your own work and the situations you encounter and write in a way that you would explain it to a neighbour. Start to think about patterns of interactions inside and outside of your organisation – micro, meso, and macro. But keep it concrete. How do things influence each other at technical, individual, team, organisational, regulatory, governmental, media, and economic levels, to create patterns and associated wanted and unwanted outcomes? Put the concepts that you read about into the context of your practice and experience of the systems that you are a part of, or interact with. The concepts you encounter will make sense not only from the points of view of what you observe in others’ work, but in what you experience in your own. Keep it short and snappy. Think short vignettes, not a treatise. Sketch out the images that come to mind (e.g., rich pictures) and start to map out some influences that you come across. Remember, thinking is more important than method, and should always precede it.
Idea 6: Observe
Arrange to observe ordinary work. It is best to observe work that you are not intimately involved with, but that you can understand well enough to know what’s going on. This might be another hospital or ward, or another air traffic control room or sector, for instance. It is essential is that you have the right attitude – apprentice, not master. It is also essential that the people you are observing consent, and understand the purpose of the observation. If you have another role that may conflict with learning how things work (e.g., competency assessor) then you have some work to do to deconflict these roles and the mindsets and perceptions that may be associated. Don’t go with a checklist. Just hang out. Notice how people resolve the dilemmas created by goal conflicts, what trade-offs and compromises are necessary, how people work around a degraded environment (staffing and competency gaps, equipment problems, procedural complexity, etc), and how – despite the context – things work reasonably well most of the time.
Idea 7: Design
At this point, you may well have ideas about improving the system structure and patterns of system behaviour (including work), to help create the conditions for success to emerge. This effort will always start with understanding the system. You’ll need to understand interactions between people, their activities, their tools, and the contexts of work (micro, meso and macro). It is advisable to avoid major initiatives and ‘campaigns’. Small designed interventions are a good way forward. You may wish, for instance to: a) make small changes to work-as-done that help balance multiple goals; b) review procedures to remove or reconcile those that are problematic (e.g., conflicting, defunct, over-specified); c) help managers and support staff to become familiar with how the work works; d) adjust buffers or margins for performance; e) review onerous analyses of events could be better directed at patterns (e.g., onerous safety analysis of multiple events outside of one’s control); f) create a means of getting regular outside perspectives on your work (perhaps an observer swap arrangement); g) create a means to simulate unusual circumstances and allow experimental performance (not a competency check). The interventions may aim at reducing unhelpful gaps between the varieties of human work (e.g., the ignorance and fantasy, taboo, PR and subterfuge, defunct archetypes). After designing, iterate to the previous ideas.