PMGT5891 Chap.11 Resilience, Bow Ties and Learning from Failure
Resilience, Bow Ties and Learning from Failure
This chapter covers the Week 10 material on resilience: how an organisation prepares for, absorbs, recovers from and adapts to adverse events once prevention has not been enough.
It explains bow tie analysis as a single drawing that links threats, preventive barriers, the top event, recovery barriers and consequences, and shows why the unit treats barriers before the knot as a measure of vulnerability and those after it as a measure of resilience.
You also meet the Swiss cheese model, tests of barrier quality, root cause analysis through the I-35W bridge cause map, and three failure cases the unit names: Comair, Sampoong and BP. A short section on the safety paradox from Greg Ip's Foolproof closes the loop.
What this chapter covers
- 01
Resilience as prepare, absorb, recover and adapt
- 02
How resilience complements risk treatment
- 03
Bow tie structure and the knot
- 04
Vulnerability versus resilience barriers
- 05
Swiss cheese model and barrier permeability
- 06
Testing barrier quality: hard, soft, independent, verifiable
- 07
Root cause analysis with a cause map (I-35W)
- 08
Failure cases: Comair, Sampoong, Texas City and Deepwater Horizon
- 09
The safety paradox from Foolproof
Worked example · free
Sorting and stress-testing barriers for a charging bay fire
- +1Preventive barriers sit left of the knot, one per threat: the charger test answers the charger fault, the driver check answers the damaged cell, and the sign answers blocked ventilation.
- +1Recovery barriers sit right of the knot because they act after the fire starts: the heat detector alarm, the sprinklers and the two-metre clear zone that limits spread to stock.
- +1Count: 3 preventive and 3 recovery barriers. Each threat line carries exactly 1 barrier, so every threat has a single point of failure on the prevention side.
- +2Weakest barrier: the sign. It has no function you can verify, nobody checks it, and it relies entirely on people noticing, so it fails the rule that a barrier must do something and be testable.
- +1Improvement: replace the sign with an airflow sensor interlocked to the chargers so charging stops when the vents are blocked, and log each trip as near miss data.
Key terms
- Resilience
- Resilience is the capacity of a system to get ready for a shock, take the hit, recover and then adapt. In a bow tie it is shown by the barriers to the right of the knot.
- Bow tie analysis
- Bow tie analysis is a diagram method that joins threats and consequences through a single top event, with preventive barriers on the left and recovery measures on the right.
- Knot
- The knot is the centre of a bow tie, the moment at which control over the hazard is lost.
- Escalation factor
- An escalation factor is a condition that weakens a barrier, such as poor maintenance, and it may need its own control.
- Root cause analysis
- Root cause analysis is a method that traces an outcome back through evidence-based cause and effect links until it reaches causes that can be corrected.
Resilience, Bow Ties and Learning from Failure FAQ
What is the difference between vulnerability and resilience in a bow tie?
Barriers on the left of the knot exist to stop control being lost, so their strength describes how vulnerable the system is. Barriers on the right act once the event has happened and limit the damage, like air bags in a crash, so their effectiveness describes how resilient the system is. A sound bow tie needs credible barriers on both sides.
Why does a bow tie usually leave out probabilities?
The unit presents the bow tie as a communication tool: removing numbers keeps attention on what each barrier actually does and whether it can be verified. Probabilities and frequencies can still be analysed with other techniques, such as layers of protection analysis, which assigns a probability of failure on demand to each protection layer.
How does the Swiss cheese model connect to bow ties?
In the Swiss cheese model each safeguard is a slice with holes, meaning imperfections, and an accident happens when holes in several slices line up. Bow ties grew out of this idea. Recording incidents and near misses against each barrier on the bow tie builds evidence of how permeable each slice is in practice.
Is a resilience plan the same as a contingency plan?
Not quite. A contingency plan is a prepared response to a specific identified risk. Resilience is broader: it is the overall ability of the organisation to absorb a shock, recover and then adapt so it ends up better placed, including for events nobody listed in the register.
Exam move
Start by drawing one bow tie from memory with every element labelled: threats, preventive barriers, escalation factors, top event, recovery measures and consequences. Then take a case the unit uses, such as Texas City or Comair, and rebuild it in your own words, marking which barriers existed, which were missing and which failed.
For each barrier ask the Week 10 questions: what makes it effective, how would you know it has stopped working, and could one failure go unnoticed among several barriers? Practise one cause map as well: pick an original incident, write the outcome on the left and ask why at least four times, citing evidence at each step.
Finally, write two sentences explaining how resilience differs from prevention, since that distinction anchors the whole topic.
Working through Resilience, Bow Ties and Learning from Failure in PMGT5891? Sia is AskSia’s AI Project Management tutor — ask any PMGT5891 Resilience, Bow Ties and Learning from Failure question and get a clear, step-by-step explanation grounded in how PMGT5891 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
