Chapter 28 — Human factors and cyber policy: failure analysis

Lesson map

Lesson 28 published. Map, example, EX082.

flowchart TD
  N0["Learning outcomes"]
  N1["28.1 A near miss"]
  N2["28.2 Contributing factors"]
  N3["Lab — Contributing factors across five near misses"]
  N4["28.3 Improvements"]
  N5["Synthesis"]
  N0 --> N1 --> N2 --> N3 --> N4 --> N5
  N5 --> EX["Worked example"]
  EX --> Q["Exercise EX082"]

Logical drawing

flowchart LR
  N0["Learning outcomes"]
  N1["28.1 A near miss"]
  N2["28.2 Contributing factors"]
  N3["Lab — Contributing factors across five near misses"]
  N4["28.3 Improvements"]
  N5["Synthesis"]
  N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> Q["EX082"]

[Learning outcomes]
    |
    v
[28.1 A near miss]
    |
    v
[28.2 Contributing factors]
    |
    v
[Lab — Contributing factors across five near misses]
    |
    v
[28.3 Improvements]
    |
    v
[Synthesis]
    |
    v
[Worked example] --> [Exercise EX082]

Learning outcomes

You will (1) analyse a near miss without blame; (2) classify contributing factors; (3) turn analysis into improvements.

28.1 A near miss

During the year, a branch assistant nearly paid a fake invoice for book supplies. The email came from a lookalike of a real supplier's domain and asked for payment to new bank details. The assistant forwarded it to the finance officer, who noticed the changed details and phoned the supplier using the number on file. Nadia analyses this near miss. Near misses are valuable because they reveal weaknesses without causing harm, but people share them only if they trust that they will not be blamed.

28.2 Contributing factors

Nadia classifies factors into individual, task, technical and organisational. Individually, the assistant was new and unfamiliar with suppliers. The task was routine, so it did not trigger suspicion. Technically, the lookalike domain passed email checks and had no external warning. Organisationally, there was no rule that bank detail changes need verification, though the finance officer checked out of habit. The lab tallies factors across this and four other near misses recorded that year, showing that organisational factors appear most often. Each factor is noted in a table.

Lab — Contributing factors across five near misses

# Contributing factors across five near misses (synthetic).
nm = [{"individual", "technical", "organisational"}, {"task", "organisational"}, {"technical", "organisational"},
      {"individual", "organisational"}, {"task", "technical", "organisational"}]
for f in ("organisational", "technical", "individual", "task"): print(f, sum(f in s for s in nm), "of", len(nm))

Run with python3 lab_factors.py. Verified output:

organisational 5 of 5
technical 3 of 5
individual 2 of 5
task 2 of 5

28.3 Improvements

Nadia proposes improvements matched to factors: a written rule that any bank detail change must be verified by phone using a known number; an external email tag; and supplier familiarisation for new staff. She praises the finance officer's habit and recommends making it a rule rather than relying on one person. She shares the analysis with staff, removing names, to show that reporting near misses leads to useful changes. Blameless analysis encourages future reports of near misses and mistakes. Managers approve all three.

Worked example

Nadia analyses a fake invoice near miss without blame, classifies factors and proposes a verification rule and external tags.

Synthetic case. The near miss is synthetic.

Errors and limits

Factor classes overlap. Five near misses are a small set.

Exercises

• EX082 — Near miss value. Why are near misses valuable?
  Show marking criteria

  Rubric: 3 = they reveal weaknesses without harm;; 2 = learning;; 1 = vague;; 0 = none.
• EX083 — Most common class. Which factor class appeared most in the lab?
  Show marking criteria

  Rubric: 3 = organisational, in 5 of 5;; 2 = organisational;; 1 = individual;; 0 = none.
• EX084 — Habit to rule. What did Nadia do with the finance officer's habit?
  Show marking criteria

  Rubric: 3 = made it a written rule;; 2 = rule;; 1 = vague;; 0 = none.

Synthesis

Blameless near-miss analysis finds system factors and turns habits into rules.

Continue studying — chapter 28 of 40

← Chapter 27 (draft) · Course chapter index · Chapter 29 (draft) →