Opening a New Station: The Change Assessment Before the First Flight

A team of air traffic controllers and safety analysts reviewing operational changes for a new airport station in a dimly lit control room at dusk.

Opening a station is filed as a commercial decision and lands as a safety one. The network department announces a route; somewhere downstream a change assessment is opened, a matrix is filled in, a colour comes out amber, and a signature is obtained. That process is worth exactly as much as the honesty of two lists: what was identified before the first flight, and what could only be learned after it. This is how those lists get built, and what has to be watching in between.

A new station is not one change

The commonest defect in these files is to treat the destination as the object of the assessment. What actually changes is a bundle of largely independent strands, each with a different owner, a different failure mode and a different amount of evidence available beforehand: a new aerodrome and its airspace, a new handling supplier, a new fuel and de-icing arrangement, a new maintenance arrangement, a new security environment, new crew duty patterns, and a new geography for the emergency response plan.

Assess “the station” and you produce a travelogue. Assess the strands and you produce something that can be argued with — which is the point.

The strands inside a station opening, and what evidence exists before the first flight
Strand What actually changes Evidence available beforehand
Aerodrome Declared distances, obstacles, procedures and minima, rescue and firefighting category and hours, surface reporting, apron layout, lighting, works in progress Substantial — published, verifiable, the best-documented strand
Airspace and air traffic services Radar or procedural control, phraseology in practice, altimeter setting units, transition levels, navigation specification, contingency routings Partial — documents describe the design, not how it is worked on a busy morning
Ground handling A supplier, its staffing, training standard, equipment condition and familiarity with your type Partial — audits describe a company, not the shift that will meet you
Fuel and de-icing Supplier, quality control regime, into-plane procedure, seasonal availability and capacity Partial, and de-icing is frequently assessed in the wrong season
Maintenance Line capability or its absence, certifying staff, tooling, spares, the real cost of a deferral raised here Good on paper — scope and approvals are documented, response time is not
Crew Duty patterns, positioning, layover, transport, hotel, aerodrome competence classification Good for the roster arithmetic, poor for what makes the rest period real
Emergency response Who reaches the site in the first hour, in which language, with what authority Weak almost everywhere, and rarely tested before it is needed

What the safety management system actually asks for

Management of change is not a refinement of a mature safety management system; it is one of its named elements, sitting inside safety assurance in the ICAO framework and elaborated in the Safety Management Manual. The obligation is to identify changes that may affect the level of safety, to assess them, and to arrange for existing mitigations to keep performing once the change is made.

The European management system requirement carries the same duty, with a procedural question attached that must be answered early: does this change need the authority’s prior approval, or does it fall under the operator’s own agreed procedure? A new destination on its own frequently does not — but what it drags with it might. A new area of operation, an approach operation not currently approved, a navigation specification not held, a new contracted maintenance arrangement: all of those touch approvals rather than internal procedure. Answering that in week one is cheap; answering it the week before launch is not.

The trap. “Three other operators serve it, so it is fine.” Another operator’s experience is a source of hazards, not a substitute for an assessment. They fly a different type, with different minima, a different schedule, a different fuel policy and a different diversion set. Their uneventful record tells you the aerodrome is workable; it says nothing about 23:40, on the last sector of a long duty, with your handler’s smallest shift.

Scope before matrix

The first real deliverable is not a risk score. It is a boundary and an assumption register.

The boundary states what is inside this change and what is explicitly outside it. Being unable to say which is which usually means two changes have been bundled into one file, which is how a new duty pattern gets approved by accident under a station opening. The assumption register does the harder work: it writes down what everyone is quietly relying on. That there will be a mechanic on the field. That the fire category holds across the planned schedule. That the handler has pushback experience on your type. That the fuel truck is there at the top of the morning bank.

Every one is testable, and every one becomes a hazard the moment it turns out to be untrue. The register is also what makes the file reviewable a year later, when its author has moved on and nobody remembers what “amber” meant.

Hazard identification then works best from mixed sources: your own event history at comparable stations, the published information for this one, and a session with the people who will operate it — a line captain, a purser, a duty engineer, a dispatcher. The workshop that produces uncomfortable findings is the one worth holding.

Before you can assess risks, you must first clearly define what’s changing and what assumptions you’re making. If you don’t, you might accidentally approve something dangerous without realizing it.

An airline opening a new station in a cold climate might assume de-icing services are available year-round. But if the supplier only operates during peak winter months, flights in early spring could face delays or safety risks. This assumption should be documented and verified before the first flight.

The aerodrome: what is published, and what is only true today

This strand is the best documented, and therefore the one most likely to be read once and considered closed. The published data is reliable; its stability is not.

Rescue and firefighting category is the classic example. The category published in the aeronautical information publication is what the aerodrome undertakes to provide; it can be reduced for training, for equipment unserviceability, or simply outside published hours, and the reduction reaches the crew as a notice they are expected to catch at the top of descent. If your schedule sits at the edge of the aerodrome’s hours, or your diversion plan assumes the category at 02:00, that assumption belongs in the register with a date and an owner.

The same discipline applies to declared distances against ongoing works, to procedure availability against navigation infrastructure outages, to runway condition reporting under the global reporting format, to pavement bearing strength against your operating weights, and to the obstacle environment against your engine-out departure procedure — the item most likely to have come from a supplier working off a data set of unstated age. Somebody has to own the question of whether that routing is current, and a station opening is an expensive place to find out nobody does.

The handler is the largest single strand

Ground damage is among the most frequent and most expensive events in an airline’s operation, and a new handling arrangement resets every assumption about it. The industry has standardised both the procedures and the audit: IATA’s ground operations manual and airport handling standards, and the audit programme built on them, are the common language for describing a handler’s capability.

The contractual layer matters more than it looks. A standard ground handling agreement’s main body is generic; the local annex is where the actual scope, equipment and local arrangements are recorded. A change assessment that has not read that annex has assessed a company rather than a station. The questions that separate a real assessment from a paper one are narrow: who trains the staff on your type, who signs off the loading, how the load sheet is produced and by whom, what happens when the ramp shift is short, how damage is reported, and what your crew does when the handler’s procedure and your operations manual disagree.

The technical strand: what a defect costs here

Away from a maintenance base, an aeroplane’s tolerance for defects changes even though the minimum equipment list does not. An item routinely deferred at base because the part is on the shelf becomes a decision about a night stop when the shelf is two thousand kilometres away. That is an operational question, not a maintenance one, and it belongs in the change file: which deferrals are realistically supportable here, what the line capability covers, who holds certifying privileges, whether tooling and a towbar are on the field, and what the recovery plan is when none of that exists.

The associated hazard is well documented: pressure on a technical decision, applied by the cost of the alternative. A change assessment that has worked out the recovery route is what protects the person making that call at midnight.

The crew strand: competence, and the first arrival in the dark

Two mechanisms carry this strand and both are frequently under-specified.

  • The first is aerodrome competence. Operators classify aerodromes according to the additional preparation they demand, the most demanding class calling for a formal briefing and, in some cases, a visit before a commander may operate there. That classification is part of the change, not an administrative consequence of it, and it has to be settled early enough for the training department to build the briefing material it implies.
  • The second is fatigue. A new station rarely leaves the roster untouched: it creates a pattern, and the pattern is what must be assessed, not the sector. Where does it sit in the duty — the fourth landing of the day, at night, at an aerodrome nobody has seen? How long is the transfer, and is the hotel quiet and dark at the hours the crew will be in it? Is there a positioning leg that turns a legal duty into an exhausting one? Transport time and hotel quality are not welfare items here; they are the difference between a rest period on paper and a rested crew.

What cannot be known before the first flight

Being explicit here separates a mature change file from a merely confident one. Some hazards are invisible until the operation exists, and pretending otherwise moves them out of the monitoring plan — the one place they could have been caught.

Knowable in advance, and only knowable afterwards
Knowable before the first flight Only knowable after
Declared distances, obstacle data, procedures, minima, published rescue and firefighting category How often the category is actually reduced, and how that is notified
The handler’s certifications, contract scope and audit status Staffing at the real peak, equipment condition on a wet night, whether your type is familiar
Air traffic services design, navigation specification, radar coverage on paper Working phraseology, non-standard local practice, readback quality on a busy frequency
Taxi routing and apron layout from the charts Which junction is confusing at night in rain, and where the runway incursion risk actually sits
Contracted line maintenance scope Response time, spares reality, and the true cost of a deferral raised here
Fuel supplier and quality regime Uplift discrepancies, truck availability at bank times, into-plane delays

Monitoring is the mitigation for everything you could not know

If a hazard cannot be closed before the first flight, the honest mitigation is a measurement with a name, an owner and a threshold that triggers action. Indicators invented after the operation has started are chosen to be comfortable; indicators defined in the change file, before anyone has an interest in the answer, are not.

Indicators worth defining before the station opens
What to watch Why it is the right signal Typical trigger for action
Ground damage and near-miss events on the ramp The single most likely adverse outcome of a new handling arrangement Any event at all during the initial period — the first one is information, not noise
Unstable approaches and go-arounds at this aerodrome Detects a procedure, runway or air traffic practice that does not fit the way you fly A rate materially above the network average for comparable sectors
Delayed departures attributed to handling, fuelling or load control A proxy for pressure on the turnaround, and therefore on everything in it A persistent pattern rather than an isolated day
Load sheet errors and last-minute changes The most direct measure of whether load control has transferred competently Any discrepancy affecting the mass and balance case
Air traffic-related events: level busts, frequency confusion, readback errors Detects the gap between airspace design and airspace practice Two or more comparable events in the initial period
Technical deferrals raised at the station and their closure time Measures the maintenance assumption directly Any deferral that could not be supported as assumed
Crew reports mentioning the layover, transport or rest The only early warning available on the fatigue strand Any recurring theme, however informally reported

The right reflex. Define an enhanced surveillance period with an end date, not an open-ended intention to “keep an eye on it”: a fixed number of rotations or weeks, a named owner, a scheduled review, and a report that says either “closing, monitoring returns to normal” or “not closing, here is what is still open”. A change that is never formally closed was never managed — only started.

Closing the file, and what should reopen it

Closure is a decision supported by the indicators, not a date that arrives. Record it, say which assumptions turned out wrong, and push those corrections back into the template so the next opening starts from a better list. That feedback loop is the only reason to keep the file.

Three men in suits review a flight path diagram on a computer monitor and a large paper chart during a meeting in an office.
Thorough planning ensures safe and efficient flight operations.

Write the reopening triggers at the same time, because the conditions that made the assessment valid do not hold indefinitely. A change of handler or subcontractor. A change of aerodrome operator, or major works. A schedule change that moves the operation into the night, to the edge of the aerodrome’s hours, or past the point where the fire category is guaranteed. A first winter at a station opened in June, because the de-icing arrangement was assessed as a contract clause and never as an operation. And an accumulation of small reports that each cleared the threshold and together describe something the file did not anticipate.

None of this is exotic: it is the ordinary discipline of writing down what you are relying on, admitting what you cannot yet know, and arranging for something to be watching in between. How other operators structure their change files is a standing topic on the FL330 forums; the process that binds you is the one in your own management system manual.

Commentaires

Leave a Reply

Découvrez l’écosystème FL330

FL330, c’est toute l’aviation professionnelle — explorez nos autres espaces.