Aviation Emergency Response Plan: Manual, Checklist, Drill
01What an emergency response plan is in an SMS
An emergency response plan, usually shortened to ERP, sets out who does what when an accident, a serious incident or another aviation emergency happens: who holds authority, who is called and in what order, what must be preserved, who looks after the people affected, and how the organization gets back to normal operations. It is written on a quiet day for an hour that will not be quiet.
In the ICAO Annex 19 framework it is one of the twelve elements of a safety management system: coordination of emergency response planning, in the first component, safety policy and objectives. The wording matters. The element is about coordination: a service provider that is required to hold an emergency response plan has to make sure it is coordinated with the plans of the organizations it interfaces with when it delivers its services, such as the airport, the ground handler, the maintenance provider and the charter customer.
The FAA says the same thing in 14 CFR 5.27. Where emergency response procedures are necessary, anyone required to have a safety management system under Part 5 must develop an emergency response plan, and the accountable executive must approve it as part of the safety policy. The plan must address at least three things: the delegation of emergency authority throughout the organization, the assignment of employee responsibilities during the emergency, and coordination with the emergency response plans of the other organizations it must interface with. Since the 2024 final rule extended Part 5 to all Part 135 certificate holders and to air tour operators under 91.147, with compliance due no later than 28 May 2027, that paragraph now applies to many small operators for the first time.
Three items is a floor, not a plan. Authority, responsibilities and coordination tell you what the plan must answer. The rest of this article is about what a plan that people can actually use looks like, and how to find out whether yours is one.
02What the ERP manual contains
A good ERP manual is short. Nobody reads forty pages of policy in the first hour, so the manual carries the decisions that were made in advance and points to the checklists that will be worked on the day. The table is a sample structure. The sections are the ones most plans need; the owners are a sample, and yours will follow your own organization chart.
| Section | What it holds | Who owns it (sample) |
|---|---|---|
| Scope and activation | Which events activate the plan, who may declare an emergency, and how the declaration is made and recorded. | Accountable executive |
| Emergency authority | The delegation of authority: who commands the response, who deputizes, and what each may decide without asking. | Accountable executive |
| Crisis team and roles | Each response function, the person who holds it, an alternate, and how both are reached at any hour. | Emergency response coordinator |
| Notification and call-out | Who is called and in what order: the authorities, senior management, the insurer, the customer. | Operations control or the duty manager |
| Response checklists | One checklist per event type, each task with a responsible role and the tasks it waits on. | Safety manager |
| Evidence and records | Preserving the recorders, the aircraft, and the maintenance, training and dispatch records; keeping the decision log. | Safety manager with maintenance |
| Families, crew and staff | Family assistance, care for the crew and for employees, one point of contact, and one spokesperson. | People lead and communications |
| Coordination with others | The plans of the airport, the handler, the maintenance provider and the customer, with a named contact in each. | Emergency response coordinator |
| Stand-down and recovery | Who may end the response, what must be on record first, and who owns the return to normal operations. | Crisis commander |
| Drills and review | The scenarios, the exercise schedule, how findings are recorded, and when the plan is revised. | Safety manager |
Two habits keep the manual alive. First, treat it as a controlled document with a revision history and an approval, the same way you treat the SMS manual it interfaces with. Second, keep the contact list as a separate appendix with its own, faster revision cycle. Names and phone numbers go stale in weeks; the delegation of authority should not change that often.
03A first-hour checklist
The first hour is where a plan either works or turns into a group chat. Several people are reacting at once, the facts are still moving, and the steps that cannot be undone (disturbing evidence, an unapproved public statement, a notification nobody made) all sit in this window. The checklist below is a sample sequence you can adapt. The owners are a sample, and the real deadlines for notifying the authority come from your state's rules, not from this table.
| Step | Owner (sample) | What goes on the record |
|---|---|---|
| 1. Confirm the facts: aircraft, flight, location, persons on board, and the source of the report. | Duty manager | The UTC time of the first report and who received it. |
| 2. Declare the emergency and pick the event type. | The person with declaring authority | The declaration, its time and the event type. |
| 3. Open the decision log. | Crisis commander | Every decision from here on, with its time and who made it. |
| 4. Mobilize the crisis team, using alternates for anyone who does not answer. | Crisis commander | Who was called, who answered, and when. |
| 5. Notify the state authority. | The role your plan names | The time, the person reached and what was reported. |
| 6. Preserve the evidence: the recorders, the aircraft, and the maintenance, training and dispatch records. | Maintenance and safety | What was secured, by whom and when. |
| 7. Account for the persons on board and open the family assistance contact point. | People lead | The manifest used and the contact point opened. |
| 8. Brief one spokesperson and hold every other statement. | Communications | The approved statement and its time. |
| 9. Call the organizations your plan is coordinated with: the airport, the handler, the customer. | Emergency response coordinator | Who was reached in each. |
| 10. Raise the occurrence record and open the investigation. | Safety manager | The two references, linked to the emergency. |
Two things make a list like this hold under pressure. Every step belongs to a role, not to “someone”, so the question “has anyone called the authority?” has one person who can answer it. And the order is deliberate: the recorders and records are secured before anything is disturbed, and no statement goes out before the persons on board are accounted for. A printed list cannot enforce that order, which is the argument for a task board that sequences the response instead of a binder.
04How to run a drill
A plan that has never been exercised is a hypothesis. There are three ways to test it, and they answer different questions.
- Tabletop. The crisis team sits in a room. A facilitator reads a scenario, adds complications, and each person says what they would do and checks it against the plan. Nothing is deployed. It tests decisions and gaps in the text, and it takes a couple of hours.
- Functional. People actually perform one function on the real phones and systems: the call-out, the notification, the records quarantine. It tests whether the function works, not whether people can describe it.
- Full-scale. The whole response runs in real time, with the airport, the handler or other outside organizations taking part. It tests coordination, which is the word both the ICAO element and 5.27 are built on, and it takes weeks to prepare.
Most small operators should run tabletops often and a functional exercise of the call-out at least as often as the contact list changes. Whatever the format, the method is the same:
- Pick one scenario from the list your plan declares, and write one or two objectives. “Can we reach every crisis team member or an alternate?” is an objective. “Test the ERP” is not.
- Name a facilitator who runs the scenario and an observer who does not play and only takes notes with times.
- Use the actual people whose names are on the plan, and the real contact list. Dial the numbers.
- Add complications the plan did not expect: the commander is on the aircraft, the event is at an outstation at night, the first report comes from social media.
- Stop on time and debrief while everyone is still in the room.
05What to record, and how findings become actions
A drill that leaves no record did not happen, as far as an auditor or your own successor is concerned. Record at least this:
- The scenario, the type of exercise and the date.
- Who took part, in which role, and who could not be reached.
- The timeline: each complication, the response and its time.
- What worked, in one line each, so it is kept in the next revision.
- The findings: every point where the plan was wrong, silent or not followed.
Then treat each finding the way you would treat a finding from an audit. A wrong phone number is a correction made the same day. A role with no alternate, a handler whose plan contradicts yours, or a team member who has never seen the checklist is a corrective action with an owner and a due date, a revision to the manual, or a training assignment. The action stays open until someone has verified that it worked, which usually means the next drill of the same scenario. That is the same discipline as closing a corrective action only on evidence, applied to the plan itself.
Finally, measure readiness by scenario. List the scenarios the plan declares and count how many were exercised in the last year. That one figure tells a safety review board more than “annual drill completed”.
06Common failures
The same handful of problems turn up in most first drills. None of them is exotic, and each one is cheap to fix before the day and expensive on it.
- A stale contact list. The person named left the company, or the number is a desk phone in a closed office.
- Authority that depends on one person. The plan works only if the accountable executive is reachable, and says nothing for the case where they are on the aircraft.
- Coordination on paper only. The manual cites the airport's and the handler's plans, and nobody has read them or met the named contact.
- A plan nobody can reach. It lives on an office server or in a binder in the building that is the emergency.
- The same easy drill every year. One scenario, the same players, office hours, and no complication.
- Findings that go nowhere. The debrief is a good conversation, and the same gap is found again twelve months later.
- No stand-down. The response fades out instead of ending, and open tasks belong to nobody.
Software fixes some of these and not others. A tool can deploy the checklist, hold the roster and keep the log; it cannot make a handler pick up the phone or replace walking the plan with the people on it. If you want to see how the Emergency Response module handles the parts a tool can carry, that page lists them.
Questions
An aviation emergency response plan (ERP) is the document that sets out who does what when an accident, a serious incident or another aviation emergency happens: who holds authority, who is called and in what order, what evidence is preserved, who looks after the people affected, and how the organization returns to normal operations. In the ICAO Annex 19 framework it is one of the twelve elements of a safety management system, coordination of emergency response planning, and it sits in the first component, safety policy and objectives.
Yes, where emergency response procedures are necessary. Under 14 CFR 5.27, any person required to have a safety management system under Part 5 must develop an emergency response plan, and the accountable executive must approve it as part of the safety policy. The plan must address at least three things: the delegation of emergency authority throughout the organization, the assignment of employee responsibilities during the emergency, and coordination with the emergency response plans of the other organizations it must interface with when providing its services.
A usable ERP manual holds ten things: scope and activation, the delegation of emergency authority, the crisis team with its roles and alternates, the notification and call-out order, one response checklist per event type, evidence and record preservation, care for families, crew and staff with a single spokesperson, coordination with the plans of other organizations, the stand-down and recovery rules, and the drill schedule with its review cycle. Each section names an owner. The contact list is an appendix revised more often than the manual itself.
A tabletop drill is a talk-through: the crisis team sits in a room, a facilitator reads a scenario and adds complications, and each person says what they would do and reads it against the plan. Nothing is deployed. A functional exercise makes people actually perform one function, such as the call-out or the notification, on the real phones and systems. A full-scale exercise runs the whole response in real time with outside organizations taking part. Tabletops are cheap and test decisions; functional and full-scale exercises test whether the plan works when someone has to do it.
14 CFR 5.27 does not set a drill frequency, so the number comes from your authority, your own manual or the standard you are audited against. A practical rule is to judge readiness by scenario, not by calendar: list the scenarios your plan declares, and check how many of them were exercised in the last year. A plan with six scenarios and one annual tabletop on the same accident scenario has five scenarios nobody has ever walked through.