The Night Order: a manufacturing safety simulation

What the group does

Over this session we’ll sit in three seats in turn: the shift team leader, the experienced line operator, and the night technician who covers two sites.

1. An ordinary night. Nothing goes wrong. We see how the shift normally runs: the workarounds, the keys on lanyards and the detector that cries wolf.
2. The night of the order. Your choices change what happens, and some of what they change doesn’t show until later.
3. The ending you reached, the world you were in, and what was there all along.

It needs a screen everyone can see and about 55 minutes. A facilitator is optional, and nobody needs to sign in or know the field.

What the debrief opens up

Its sections:

• The quick way
• What the text did
• Same choice, different night

Some of the questions it asks the group:

• The stop at ten past two cost £65k in easy mode and prevented a recall in hard mode. Same decision. How does our organisation record each?
• The override key lived on the team leader’s lanyard. Who put it there, and what did that say about what the site expected?
• What did the half past one text ask for, and what did it make happen?
• Where in our work does a number on a board influence what gets said?

Same choices, different ending

The group plays in one of two modes, picked at the start or at random. The same choices can end differently in each, which is the point: outcomes are rarely ours to choose.

• Hard mode. The depositor guard’s hinge bolt is worn. Lifted a third time, or run over speed, it sheds a fragment.
• Easy mode. The hinge is fine. The reject at 02:10 is the detector’s own fault, like the two earlier this week.

To see the debrief a group reaches, without running a session, open the debrief preview.

Further reading

• Stopping the line. Toyota thanks the person who pulls the cord, and treats false alarms as useful drills. Without that, the cord doesn’t get pulled. The Andon Cord, Tom Geraghty.
• Production pressure. The order, the lorry at six, the fine and the site manager’s text all pushed the same way, though nobody told anyone to skip the procedure. The next morning, the night is judged in money: stopping the line shows up as a cost, and a near miss that nobody logged doesn’t show up at all. The Amagasaki Disaster, Tom Geraghty.
• Normalisation of deviance. Workarounds become normal because they were fine last time. Deviance and adaptation are also why most organisations work well day to day. The Challenger Disaster: Normalisation of Deviance, Tom Geraghty.
• Public, private and secret. Which workarounds we tell a manager about, which we tell a trusted colleague, and which we tell nobody. “Much of psychological safety is about widening the private space and shrinking the secret one.” Work as Imagined vs Work as Done, Tom Geraghty.
• How we respond matters. Speaking up depends less on what people say than on how they predict they’ll be received, including by a tired manager in the middle of the night. How We Respond Matters, Jade Garratt.
• Safety-II. Things go right and wrong for the same basic reasons, which is why the ordinary night matters as much as the incident. Safety I & Safety II, Tom Geraghty.
• Weak signals. A guard “getting loose”: the kind of signal that is only obvious afterwards. Amplifying Weak Signals, Tom Geraghty.

Key papers

• Migration towards the boundary. The model behind the drift diagram: pressure towards efficiency and the gradient towards least effort move work towards a boundary of acceptable performance that nobody can see until it is crossed. Rasmussen (1997), Risk Management in a Dynamic Society: A Modelling Problem, on the Psych Safety field guide.
• Workarounds instead of fixes. Frontline staff work around failures to get the job done, and the problem that caused them is never fixed. Time pressure and the cost of escalating keep it that way. Tucker & Edmondson (2003), Why Hospitals Don’t Learn from Failures, on the Psych Safety field guide.
• Normalisation of deviance. How a deviation that causes no harm becomes the new normal, through sensible decisions made one at a time. Vaughan (1996), The Challenger Launch Decision, on the Psych Safety field guide.
• Hindsight and outcome. Error as the starting point of an investigation rather than its conclusion, and how knowing the outcome changes how we judge the same decision. Woods et al. (1994), Behind Human Error: Cognitive Systems, Computers, and Hindsight, on the Psych Safety field guide.
• Ten principles for systems thinking. Including demand and pressure, trade-offs and local rationality, written for the people who run systems rather than for academics. Shorrock et al. (2014), Systems Thinking for Safety: Ten Principles, on the Psych Safety field guide.

Sources

• Hollnagel, E. (2009). The ETTO Principle: Efficiency-Thoroughness Trade-Off. Ashgate.
• Dekker, S. (2011). Drift into Failure. Ashgate.
• Conklin, T. (2012). Pre-Accident Investigations. CRC Press.
• Garratt, J. (2026). Learning Teams. psychsafety.com.