Observations
Watch normal work on the ramp, note what you see, and code it afterwards.
See what goes right and what goes wrong
Watch a normal turnaround from arrival to pushback, say what you see, and tap Managed, Mismanaged or Unmanaged as it happens. No names, no blame: it's about the work, not the people.
| Ref | Date | Turnaround | Notes | Status |
|---|
No observations yet. Start one before the aircraft arrives.
New observation
Fill this in before the aircraft arrives. Don't record names, staff numbers, flight numbers or registrations.
Notes
Say what you see: what and how, not why. Tap the microphone on your phone's keyboard to dictate. Step away from the crew to speak. Using paper? Type each note in with the time you wrote it.
Notes so far
No notes yet.
Code
Select words in the notes, then choose a code. Code what is there. Links between codes come from the analysis later, so don't try to connect them here.
No notes to code yet.
Codes
Checking
Read the notes against the codes. Change codes above when the group agrees. If you can't agree yet, write the question here and come back to it.
Set up
Settings for this device.
Check the coding together
The checking group (2 or 3 observers) reads each set of notes against its codes and agrees any changes. Only checked observations count in the results.
| Ref | Date | Codes | Flagged for a second look | Status |
|---|
Nothing to check yet. Observations appear here once they're coded.
Results
From checked observations only. These are counts and links, not proof of cause: use them to choose what to look at with the people who do the work.
No checked observations yet. Results appear once the checking group has marked observations as checked.
Round by round
Turnarounds with at least one of each, in each round. Use it to see whether changes made a difference.
| Round | Checked | Threats | Variations | Undesired states | Enablers | Good practice |
|---|
Summary
How the turnarounds ended: safety and quality
Operational outcome: on-time departure
Operational, not safetyKept separate from safety on purpose: on-time is never combined with safety outcomes, and never shown by crew, shift or observer. It's about the conditions the team works in, not about people.
How recovery happens
When teams managed a threat, variation or undesired state, what enablers and good practice were there?
How often each kind of code turned up
Compared with the reference
The reference (Round 0) is the NOMIC reference data: 1,302 turnarounds observed across several airlines in one airline group's network. It's a point of comparison, not a standard to meet. More variations can mean the observers are seeing more, not that work is less safe.
About the reference
Links between conditions and what happened
Which conditions go with more (or fewer) variations or undesired states in the same turnaround. Odds ratio with 95% interval, the same method as the PhD research, with the same link in the reference where there is one. A link is not proof of cause: it's a question to take to the people who do the work.
Expand: what goes with work going well
Linked to the turnaround going well, or to fewer variations or undesired states. Things to have more of.
Shrink: what goes with work getting harder
Things that make the work harder. Look at the conditions, not at the crew.
For a 60 to 90 minute session with people who plan the work and people who do it. Pick one topic per session.
What the data shows
From the notes
Questions for the group
Actions to re-measure next round
| Action | Person responsible (role) | Check |
|---|---|---|
| Next round | ||
| Next round |
Sign in to code and see results
Recording observations works for everyone. Coding, checking, results, the Learning Team pack and observer practice use the full coding guide and the reference data, which are for approved airlines, ground handlers and airports.
Not a member yet? Ask for access on the same page. Once you've signed in, come back here and reload.
On the ramp
Watch a normal turnaround from arrival to pushback, say what you see, and tap as it happens.
Before the aircraft arrives
- Introduce yourself to the loading supervisor (the leading hand, crew chief or team leader, wherever you work). Say you're observing normal work: no names, no blame. It's never used for discipline, and managers only ever see totals, never a single turnaround.
- Wear your hi-vis and hearing protection. Stand where you're safe and out of the way.
- If the crew would rather not be watched, thank them and walk away. No questions asked.
- Fill in the turnaround details, then start your notes.
Where to stand
- Outside the safety zones: clear of engine and propeller areas while the beacon is on, the pushback path and equipment lanes.
- Don't help with the work, and follow your airport's rules for using a phone airside.
- Too noisy to dictate? Step away to speak, or jot notes on paper and type them in afterwards with their times.
What to say into the phone
- What and how, not why. Say what people do and what happens, in plain words. Don't guess reasons.
- In Preparation, describe the scene: weather, noise, congestion, who's around. After that, only what changes.
- Step away from the crew to speak. Each note keeps its time.
- Four phases: Preparation, Unloading, Loading, Departure.
What to look for
- Threats: things that come at the team from outside: late freight, a broken loader, unclear information, a bay that isn't ready.
- Variations: where the work differs from how it's planned or written down.
- Undesired states: where the safety margin has gone: a load not secured, a near collision.
- Enablers and good practice: what helps the work go well, and skilled work worth learning from.
The three taps
- Managed: the team acted and it was resolved.
- Mismanaged: the team acted but it wasn't resolved, or made it worse.
- Unmanaged: no action was seen.
Tap while you talk. The tap is timed to the second and goes with the note you're saying.
If something is unsafe
Safety comes first. If someone is about to be hurt or the aircraft damaged, stop it, as you would on any shift. Note that the observation stopped early.
If the crew ask
"I'm watching normal work to see what makes the job harder and what helps. No names, never used for discipline, and managers only see totals. The results come back to the crews first."
When the aircraft has gone
- Answer the questions about how the turnaround ended: on time, loaded correctly, any damage or injury.
- Tap Finish and send. Once your organisation has it, it is cleared from this phone.
Dashboard
The latest round at a glance, from checked observations only, and where you sit against the reference data and other organisations.
Normal Operations Monitoring: summary
Totals and links from checked observations. No notes and no single observations. Links in the data, not proof of cause.
Action plan
What each Learning Team decided to change, who is responsible, and one measure to watch. The next round shows whether the change made the job easier.
Kept on this computer and in backup files.
Code this turnaround, then compare
Code the narrative the way you would a real observation. Then see the agreed answer from experienced observers and how close you were. Your scores stay on this device and never appear in reports.
My codes
How close you were
Your scores stay on this device| Agreed answer | Your code | Result | Management |
|---|
Select a row to see the words in the narrative. Match: same code. Close: a different code in the same category, counted as half. Missed: in the agreed answer, not in yours. Extra: in yours, not in the agreed answer. Score = 2 × matches ÷ (agreed codes + your codes).
Towards the certificate
Agreement on categories uses Cohen's kappa: for each exercise and each category, did you and the agreed answer both find it, or both not? Only categories that turn up in either coding are counted, and kappa takes out the agreement you'd get by chance. 0.60 or more is usually called substantial agreement.
You've reached the pass mark
Print your certificate of completion, or save it as a PDF from the print window.
Back up
Observations live in this browser on this device. Clearing the browser's data, or losing the phone, loses them. Save a backup file often and keep it somewhere only the observation team can open.
Save a backup
Everything, including the narratives. Treat the file as confidential.
Load a backup
Adds the observations in the file to this device. Use it to restore a backup, or to combine another observer's work. Where the same observation is in both, the more recently changed copy is kept.
Spreadsheet export
For Excel, Power BI or your safety system. Observations: one row per turnaround. Codes: one row per code, with the words coded. Observations use neutral numbers (e.g. OBS-0142), never observer initials; the Observations list shows which working number each one matches. Dates are given as the month and there are no times, so a row can't be matched to a roster. They still include station and handling company: keep these files inside the programme team, and never match them to rosters or flight records.