A Possible answer to Runway Incursions and similar tragedies

Background 

What we have now at major airports is a highly complex and confusing taxiing "road system" with confusing signage, no traffic cops and very few traffic lights. The traffic gets quite heavy at times, human errors are inevitable and you can end up on a racetrack headed the wrong way.  What is required is a system that can, as far as possible, rule out human error actually causing an accident.

Proposal

If each runway take-off entry or intermediate taxi-in crossing-point intersection had a holding-point treadle (weight-actuated) incursion alarm blatting out on Tower and Ground frequencies, each incipient incursion would immediately be notified (by an attention-getting warbling tone) to pilots on both the Tower and Ground Movement Frequency (for pilots in the pattern, on ILS finals, taxiing and lined up on the runway or rolling). Any taxiing pilot who thought he’d triggered that VHF audible alarm would immediately stop/check his position. In fact, all taxiing aircraft and those about to roll would be mandated to do so (i.e., stop and check for the hazard aircraft or vehicle). An aircraft taking off in poor visibility (below decision speed) would abort.

Vehicular traffic (sweepers, follow-me vehicles, fire trucks, etc.) would also trigger alerts and themselves be warned immediately of their infractions (working on the principle that all authorized vehicles will be radio-equipped and maintaining a listening watch on Ground and/or Tower).

The treadles, twin pressure-sensitive transducers a few feet apart (and similar to traffic counters seen everywhere on the highways), and associated logic circuitry and plumbing, are not that daunting a technological prospect. The installation could feature an adjacent nosewheel-activated strobe light . That would immediately pinpoint the perpetrator and also alert the offending crew, even in bad visibility.

If you wanted to make it even more reliable and indicative, you could have a green/amber/red light at each take-off runway entry taxiway treadle - directed at the next take-off candidate:

Red:      Treadle armed and ready to trigger.

Green:  Disarmed and clear to cross.

Amber: Unserviceable fault annunciator (dead treadle operations in progress).

Because the problem is runway incursions, you have to put the burglar alarm on the runway itself (i.e. its access points).

I can think of no other way that would ensure that the only trustworthy common denominator (airplane weight) will be able to do what the human element in the equation is demonstrably unable to reliably do. 

Quest:      When would it be disarmed for runway(s) in use?


Answer:  a. Take-off -  Immediately prior to (or as) an aircraft is being cleared to cross its treadle for line-up / take-off. In order to maintain the integrity of runway security, only the take-off access taxiway's treadle-pair would be disarmed (for aircraft entering to line up).
                 b. Landing/runway exit - No requirement to disarm (read why later)


 Quest:   When would it be armed?

Answer:  At all other times (to ensure the sanctity of the runway from incursions at any time).

              i.e. Runway protection needs to be more ON than OFF.


 Quest:  What would be the action of a pilot on take-off (on hearing the alarm)?


Answer:   Look for the possible threat and take appropriate action


 Quest:  What would be the action of a pilot on landing approach (on hearing the alarm)?


Answer:   Go around


 Quest:  What would be the action of a pilot on roll-out (on hearing the alarm)?


Answer:   Look for the threat and take a runway-side departure solution if necessary to avoid a collision.


 Quest:  a. What would the ATC operator's switch look like? b. ......Armed Indicator look like?


Answer:       a. hand-held carrier LSO's style wave-off lights switch  


 b. Light inside binocs  / Light on hand-mic / line-of-sight light on console between the controller and the runway threshold.


Quest:   When would the ATCO cancel an alarm?


Answer:  When situation resolved or demonstrably false or after 20 to 25 seconds of warble (requiring an immediate broadcast clarification of alarm cause by the ATCO). Controller would require a control console with built-in fault-monitoring and selectable treadle sets for different duty runway(s)).


 Quest:  At which airports would you propose using it?


Answer:  Any airport with half O'Hare's annual movements (or greater)

Quest:   Can you think of any other enhancements? Perhaps using the aircraft's own lighting to identify the intruder?

Answer:   Yes there are a few important ones. Look at this URL

Challenge:  PICK THE FLAW IN THIS type of anti-incursion SYSTEM (if any).

Cost?   Controller inability to operate?  Loopholes in the system?   Legal Problems?   Pilot Opposition?  Slowing of Operations?

Advantages: 1. Puts the onus upon the runway ATC controller to "safeguard" his runway for the aircraft that he's cleared (to land or take-off). And that's where the onus should be.

                    2.  A nominal "fines" system should put both pilots and the ATC operators on their qui vive [so as not to be embarrassed by causing an alarm (false or otherwise)]. I suggest the "fines" should be stockpiled in the individual airport improvements votes (rather than FAA consolidated revenue) - and be administered by SATCO's.

                    3.  If initiated by the US, it might prove its worth and eventually "shame" other nations into implementing such a system.

                    4.  The greatest advantage of this system is that it would be similar to an air raid alarm. It gets everyone on freq - including crash-fire-rescue - heads up and looking for the threat (which is now visible). Most runway incursion ground collisions were almost totally unforeseen by participants and ATC - until they had already happened.

                    5.  Low tech, based on simple technology and not a lot of maintenance.

                    6.  No airplane weight penalties for on-board equipment. Not much training required.

                    7.  If a treadle-pair goes unserviceable, a temporary one could be slid into place. Serviceability would be monitored by the Tower's console.

            8.  For a ground-radar system to work, it has to be operable 100% of the time AND the operator has to be looking at it just before the incursion takes place.

            9.  Dreadle is a lot cheaper than the aircraft-activated "follow me" lights guidance system.

and finally, most convincingly

    10.  Look at this. This is the hopeless system upon which the FAA has pinned its hopes.

 

 Cost of such a system? - A lot less than a runway collision and probably of the order of $15M to $25M per airport (depending upon size, number of parallel runways etc)

 Disadvantages?  It would need to be trialled and could only be implemented once all treadles were in place. Treadles? All USN VP Air Stations have them for actuation of their aircraft rinse-racks. They could tell you the cost (ask NAS Moffet, NAS Barbers Point or Alameda). Alternatively ask the Highways Dept.

                         The need to await aircraft taxiing in from the runway to clear an armed treadle could be avoided by having a two treadle (i.e. paired) system. Aircraft entering runways would trigger the warbler of an armed system by an A-B progression, and aircraft clearing the runway on taxi-in to the ramp would not, because they would represent a B-A progression across the treadle pair. (and this B-A inbound crossing would also present a reassuring green light to the ATCO in poor visibility).

 What's my name for the system? I call it DREADLE. It speaks volumes.

Justification

It is about time the FAA started thinking laterally and moving away from the super-techno solutions (inspired by the corporate weenies looking to cash in on a crisis). Dreadle also works most invaluably in low visibility operations, and addresses the fact that pilots busy in the cockpit during taxi (and not looking out as much as they perhaps should) would never miss its unsubtle prompt. And you cannot get more timely a warning, for all concerned, than an automated treadle-operated VHF warbler.

When an ATCO issues a clearance to land (or take-off), he is trusting normally upon his elevated vantage point to see that the runway is clear - but the incident/accident files are full of instances where it wasn't (and not just talking LAHSO operations here). When he issues that clearance he should have the assurance that comes with having Dreadle-armed that runway (so that any incursion will be heard by all concerned - and straightaway without any lethal delay). A trite point - but an interesting legal one when they hold an ATCO accountable for a runway collision (or near collision).

 And it's also well past time that the FAA effectively addressed this highest of the high hazard areas. I like low-tech (per the KISS Principle) and DREADLE is something that pilots (US or visiting / ATPL or PPL) will easily understand. A cautionary treadle-triggered warbler over the VHF first up gives rise to "Oh God, I hope that's not us, check up quickly." (for taxiing brethren). Of course for the crew on take-off and those on approach, they instantly know that something's up and will be in a better position to retrieve their situation before tragedy strikes. If you think about it, the only universal medium that puts ATC, Crash-fire-rescue and all pilots on frequency onto the alert simultaneously (with nil further correspondence required) is their commonly shared frequencies (which will be one of Ground or Tower for those in contention). Firemen always monitor TWR freq as well as Ground. It should also be mentioned that the warbler would not be any ear-splitting shriek, simply a background-identifiable superimposed and modulated tone. It could have that well known sexy voice whispering throatily (every 10 seconds) "Runway incursion alert, check your position". I suggest a five seconds on / ten seconds off transmission cycle and a low-power transmitter so that it can easily be over-transmitted if necessary (and/or cancelled by the ATCO if he sees a need to transmit) or has NB'd the cause (culprit) - and needs to transmit in order to resolve it (or announce: "Runway incursion alert resolved and cancelled at time .....").  A low-power transmitter not only makes it over-transmittable but also keeps it short-range within the family (those within about a ten mile radius that have the "need to know"). Putting it on 121.5MHZ (distress freq) wouldn't be as good an idea - simply because not all pilots monitor that. It should be on both of the two primary freqs in use - to be bullet-proof. One of the issues that would need to be resolved is the treadle trigger-weight threshold (for bug-smashers?).

 Another beauty of it (as I see it) is that ATCO's and pilots are always in a competitive environment as far as "not screwing it up" goes. The potential embarrassment factor might be sufficient to preclude the majority of incursions - which are simply due to inattention. And of course the other well-worn principle in the world of acronyms and overflowing glossaries is: "Give it a catchy name and it will be on everyone's lips - and the World will beat a path to your door".  At least I think that's how it's supposed to work. But hopefully it might set the ball rolling in a different direction and tend to concentrate that introspective FAA mindset on the true nature of the problem.

 "Aircraft on short final. Line up and hold behind the landing aircraft". Not an unusual situation at all, so to cover the question of just when the ATCO should disarm Dreadle (in order to permit this aircraft to enter the runway and line-up), I'd have to amend the landing-rollout disarm situation as being "landing aircraft's nosewheel on". That should satisfy the nit-pickers and minimize residual threat exposure. It could even be "main-wheels on" - but in any case, it would need to be closely coordinated by the ATCO in order to minimize false alarms.

 Of course the odd man out (reference that recent incursion by a chopper in the UK that caused a 757 abort) would be the helicopter. But unless it's mandated that they must always roll over the runway entry holding point on wheels, that's the only identifiable flaw in the theory (so far).

 Seeing an adjacent red light switch to green would be a backup confirmation to a crew that they now have their line-up clearance (at least). But of course it would not avoid a repeat Tenerife - except inasmuch as any aircraft backtracking to clear the runway would cause the ATCO to KEEP the runway armed (and his own red light would be a reminder that the runway was danger-ground)

 SQ006 Accident:   How would this system have prevented last October’s fatal crash of Singapore Airlines Flight SQ 006, when the pilot attempted to take off on a closed runway? Taipei ATC was using that closed runway as an exit taxi-route so they couldn't really block it off - but there was another (Dreadle) solution available. B-A progression would have worked (and allowed aircraft to exit via taxiing along an "armed" disused runway). BUT A-B treadle progression would have immediately alerted ATC (and the crew.... and also the crew that watched them do it and said nothing) that some hapless soul was lining up on the wrong runway (having just entered the out-of-use runway for line-up - thereby triggering an A-B alert). The cost of that Taipei accident alone would cover installation of Dreadle at 20 other airports. So runneth the affordability question.

The ‘Dreadle’ Concept

A and B would be the normal take-off runway entry points based on the fact that airliners do a balanced field take-off based on accelerate/stop distance available.

Regional/commuter aircraft might be directed to taxiway entry point shown at C. Whether or not positions at C are fully Dreadle-equipped (with lights) would be a matter for individual airport safety managers. Whilst it is possible that works in progress might cause taxiways (marked "D") to be used for enter/backtrack/line-up, they would be treadle-pair, but not light equipped - simply to guard against inadvertent entry.

The aircraft waiting at B may have inadvertently passed over the treadle-pair arrangement only with his nosewheel. It is suggested that this should trigger an ATC only (i.e., non-VHF) alert in the tower (audio plus flashing light). But when both nosewheel and main gear pass over the treadle-pair, a full alert should sound. In other words, an aircraft (or ground vehicle) lost in poor visibility or misunderstanding taxi/take-off clearance would cross completely over any treadle-pair with an identifiable interval between front and rear wheel sets. In poor visibility, tower staff would probably not see this, and so the event should trigger a full VHF alert. 

 If you wanted to protect entirely against the totality of "poor visibility / lost on taxi / confused with his clearance" circumstances, then ALL red and green marked runway entrance taxiway inverts (on the diagram above) would need to be Dreadle protected. That's when the cost rises. However given that Dreadle treadles need only be a two-bar arrangement across the taxiway (similar to the traffic counters that you see everywhere on the highways and byways) with a pair of pressure-sensitive transducers, a logic unit (for B-A / A-B resolution and gear-passage count) and a signal amplifier  .......well the more you have of them, the less you need pay (per unit cost).  I think it may be possible to plumb it into the present airfield-lighting cabling. And if that's not possible, at least to utilize the same ducting for Dreadle wiring.  Electronic Design Work is not one of my fortes but I think you'd have to agree that it's not really all that daunting a technological prospect. Cost would be the spectre raised by the FAA for sure (I would guess) - rather than technical complexity or desirability.  If you wished, each Dreadle permanent (or temp) installation could logically have a nosewheel activated strobe-light arrangement alongside. That would immediately pin-point the perpetrator, even in bad visibility, for partial (i.e. nosewheel only) or full (i.e. broadcast) alerts.

 In poor visibility the fact that the aircraft at A has definitely cleared the active runway (thereby allowing a take-off clearance) can be passed by  the B-A treadle progression giving the ATCO a reassuring green light  (remember that a nosewheel (only) A-B incursion would give the controller his own private alarm red light plus audio - as against a full VHF broadcast Dreadle alert). This would be on a Dreadle control and indication panel in the tower and would also accord an ongoing check of treadle serviceability). So this is obviously another case for all runway access and exit taxiways being Dreadle-equipped.

Design

Dreadle sites and arming selections would have to be "nutted out" for each airport because they each have their own design features and particular prevalence for incursions as a result of taxiway complexity and prevailing poor visibility (plus foreign pilot through-put). At present the FAA appears to be pinning its hopes to a surface radar system, however that's got a fallible, distractable human interpreting it (and it may only be his secondary duty). They are also running awareness programs - but in my view that's like praying for rain. In my opinion, if the ATCO is responsible for avoiding runway incursion accidents he needs the backup of being able to "arm" his runway against intruders. Once the visibility is way down and there's lots of traffic out there, each aircraft doing its own thing, coming perhaps from different sides of a terminal hub and different runway exit-points, it's simply unrealistic and unfair for a tower operator and surface movement controller to be "responsible" for that traffic (which can often be sight unseen in poor visibility). You only have to look at the alarming increase in the incidence of incursions to understand why it is the number one accident prevention priority - and always will be.

Summary

 Unfortunately, as long as the FAA avoids any biting of the bullet, there will always be a major incursion accident just around the next corner. Collisions always involve double the number of casualties because there will normally be two aircraft - and at least one aircraft is going to be fully fueled. That's the nature of the beast. When you look back at Tenerife and project forward to the A380 (as a participant), the cost of that potential accident justifies whatever expense needs to be outlaid - albeit only at selected airports (half O'Hare traffic density and upwards is as good a rule of thumb as any).  (However initially they should pick a suitable airfield (with an existing incursion problem) and trial an installation. In no way could it ever be one-eighth as complex (or expensive) as the introduction of a new radar system. The software for it could be written by a school-kid.

Input Commentary / Suggestions:  To the email address below please.

Comments received Why Dreadle is needed The FAA Seeks a Solution   FAA Runway Safety Program
       
Another Sad CaseNeed Dreadle?Incursion Statistics

Making Light of Dreadle

  The Dreadle Alternatives (Judge Dreadle - for yourself) 

Final Report - Sub-Committee on Runway Incursions

FAA Runway Safety Report - 1997 - 2000

(PDF File - 1.59MB)    "Please right-click and download file before viewing"

  Return to Commentary