Airport Pavement Condition Surveys for Government and State Airports

Who runs into this

An owner of several airports manages a network, not a pavement. A ministry, a regional authority or a state airport company holds a runway at one airport and an apron at the next, each with its own age, traffic and repair history. One capital plan covers them all. The question is which sections go first, and the answer has to hold up in front of an auditor and a budget committee.

Three things get in the way.

• Condition data comes from walked visual inspections. Two inspectors rate the same crack differently, and one inspector rates it differently two years apart.
• Each airport was surveyed in a different year, by a different contractor, to a different scale.
• A walked survey rates the units a person reaches and leaves the rest to be inferred.

The pavement condition assessment is the service behind this page. It is in development and offered as early access. This page sets out what it does for a public owner, and where it stops.

What the rules ask, and what they leave open

Regulation (EU) 139/2014 sets the requirements for certified aerodromes in the EU. ADR.OPS.C.010 covers pavements, other ground surfaces and drainage, and it asks for the surfaces of the movement area to be inspected and kept in a condition that supports safe operations, under a maintenance programme. ICAO Annex 14 Volume I Chapter 10 says the same thing for aerodrome maintenance, and ICAO Doc 9137 Part 9 gives the guidance behind it.

No EU rule names a Pavement Condition Index. The owner decides how the maintenance programme is evidenced, and most fill that space with a walked inspection written up in the inspector’s own words.

A PCI fills it with a number. It scores a sample unit from 0 to 100 from the distresses found on it, and a section is the area-weighted result of its units. The figure does not satisfy a regulation, because none asks for one. It gives the programme a measure a decision can rest on. The PCI explainer shows how the number is built.

Budget evidence that compares across airports

A capital plan needs a figure that means the same thing at the first airport and at the last. A walked inspection cannot promise that. The rating depends on who walks and on the day, so a comparison between two airports compares two inspectors as much as two pavements.

The drone survey takes the inspector out of the rating. The software finds each distress, measures its length in metres and its width in millimetres, and rates it by the D5340 definitions. The same rule runs on every pavement, so a change in the figure is a change in the pavement and not in the person.

A runway section at Žilina (LZZI), Slovakia. Seven sample units are coloured by PCI, the section PCI is 68.5, and every crack is drawn where it lies.

That map is the budget exhibit. A planner sees which units are worst, and an auditor sees that the figure comes from the drawn cracks.

One honest limit belongs here. The report lists every distress type it does not rate, and an unrated type biases the PCI upward. The rule is the same at every airport, so the ranking stays consistent, but the absolute figure is an upper bound where unrated distress is present.

A neutral method, not a vendor index

ASTM D5340 is a published standard. The FAA uses it for airport pavements, and the US DoD applies it through UFM 3-260-16, whose distress catalogue is aligned with it. TarmacView follows the D5340 definitions and its catalogue is transcribed from UFM 3-260-16. The PCI of the UFM 3-260-16 worked examples is reproduced within about 1 point.

For a public owner the figure is yours. Any consultant who knows the ASTM D5340 method can read a unit sheet, check the deducts and challenge the result. A later tender can ask for the same method from a different supplier, and the old and new surveys stay comparable. A vendor index makes the next tender a negotiation with the last winner.

The same rating whoever flies

A programme that spans several airports and years will see several operators and several tender rounds. The rating does not depend on any of them. The mission planner is drone agnostic, with export to DJI, MAVLink and UgCS, so the tender asks for flights and not for one aircraft brand. The software rates the images the same way, so the figure from the first round and the figure from the fifth are made by the same rule.

Wind, temperature and rain limits are the aircraft maker’s. A survey flies in daylight, preferably over a dry surface. Any asphalt or concrete not covered by snow, sand or water can be surveyed, on an airport the operator knows or one it has never seen.

The runway reopens when the drone lands

The aircraft launches from the shoulder. Nothing touches the pavement and no inspector walks the surface. Runway time is the flight itself, so the runway reopens as soon as the drone lands. One aircraft covers 1 km of a 45 m wide runway in about 4 minutes for cracks from 10 mm, 13 minutes for cracks from 5 mm, and 45 minutes for the full 1 mm per pixel survey with cracks from 2 mm, and more aircraft cut these times. The software works on the stored images after the flight, off the runway, and a 5 mm report for 1 km is ready about 20 minutes later.

Data stays with the owner

The processing runs on a MacBook or a PC with a graphics card, on your own hardware, with no cloud service in the processing path. The imagery stays with the owner unless the owner sends it.

The output is three open formats. A GeoJSON inventory holds every distress with its map position, length and width, plus the PCI per sample unit and section. A PDF condition sheet per unit shows the deduct calculation, so a reader can follow the arithmetic. CSV tables carry the numbers into the owner’s own tools. All three open in standard software, so the record moves to another system or supplier without an export step that belongs to us. The offline workflow describes an airfield with no network in full.

Analyst review keeps the figure honest

The first pass is automatic and the report never waits for review. In review, the analyst sees each detection next to the cracks that look like it. A sealed crack, joint, vegetation, paint or stain is marked in one save, and the PCI of the unit changes at once. Marking one sealed crack can clear 24 similar ones in the same save. Every verdict is saved as a labelled example for retraining.

From repair after failure to planned repair

One survey is a snapshot. Two surveys are a trend. Every distress keeps its map position, its length in metres and its width in millimetres, so a repeat survey of the same pavement shows how each crack has grown. Concrete fits its slab grid, so a slab keeps its identity between surveys.

That is what moves a maintenance plan from repair after failure to planned repair. A crack that opened by a few millimetres in a year sits on a different curve from one that did not move, and a programme that sees the growth can seal on a plan. A growth forecast is in development. What exists today is the measured growth between two surveys.

How a network survey runs

The first step sets the section map every later survey reuses, so it is the one worth getting right.

Limits that do not move

• Daylight, and preferably a dry surface. Snow, sand and standing water hide the cracks.
• Wind, temperature and rain limits are the aircraft maker’s.
• Sealant and paint edges read as cracks. An analyst clears them in review, in minutes.
• Nothing below the surface. The survey does not measure structural capacity or friction.
• Cracking and spalling are rated automatically today. A classifier labels the other D5340 types, and the report lists every type it did not rate.

Where it is going

The direction, with no dates. Every type rated automatically with severity, debris flagged in the same pass, results loading into PAVER and GIS, crack growth forecasts, crack depth from angled photos, and drone docks at the base. None of this is built, and the survey described above does not depend on it.

Objections worth raising

“No rule asks for a PCI, so why add one?” None does. The rules ask for a maintenance programme, and a PCI gives it a figure that auditors and a capital plan can use.

“Our inspectors know the pavement better than software does.” They do, and they stay in the loop, because the analyst review is theirs. The software adds the same rule on every unit and the whole surface imaged.

“We will be tied to one drone or one supplier.” The mission export covers DJI, MAVLink and UgCS, the output is GeoJSON, PDF and CSV, and the method is published.

To test it on your own sections, contact TarmacView about early access.

Related tasks

If the question is the service itself, the pavement condition assessment page describes it while it is in development. If your airfields are rated under a NATO or national military method, the NATO airfield pavement condition page takes that angle. If the owner is a US military installation, the US military airfield pavement PCI page covers the D5340 and UFM 3-260-16 rules. For the lighter recurring look, the runway pavement check is the shorter service.