The problem with runways at Canada's major airports - CBC

[Cliff Jumper]

The person(s) who thought up deicing at our major airports should be shot and pssed on. Toronto is the worst.

and then.....

even if it seems to work well there is always room for improvement. Why are people so opposed to change.

I hear you! Couldn't agree more. It works well, with room for improvement, but the designer should be urinated on, and murdered.

[cossack]

It isn't the location of the CDF that's the problem, its the location of the terminals. Crossing active runways is never ideal and that's what we are forced to do. It impacts on the movement rate and its has risks.
In my 15 years here, I've not heard of holdover issues unless its been caused by a flow delay, then the flow time gets amended.
As for the leaving CDF and departing in 2 or 3 minutes? You've got to be kidding? There is seldom a long line for departure if everyone is deicing (unless snow removal has caused a delay) because we can depart at a much faster rate than aircraft can be deiced. That's why they want to make the CDF bigger in YYZ.
The delays to departure that have become more prevalent over the last couple of years is the engine runs prior to departure. 10 seconds, 30 seconds, only on the runway after take off clearance has been issued, within 5 minutes of departure. WTF? We try very hard to accommodate but it is very labour intensive and requires a lot of radio chatter. That's the major impediment to your de-ice and go philosophy.

[complexintentions]

Margins are built in to account for rare failures, not wilful negligence. Too harsh? Ok, dismally poor decision-making and completely incorrect procedure, then.

Nope.

Human errors exactly why RESA's are built. How many overruns/undershoots were caused by 'rare failures'?

A few, but not many.

You've completely missed the point. No amount of margin will suffice if crews make multiple gross errors. Certainly, human errors are the main cause of overruns/undershoots, and that indicates a failure of training and proficiency, not of facilities. Which was, actually, my point if that's not clear.

On any modern airliner there are sophisticated performance tools that will very precisely indicate if the aircraft takeoff can be rejected safely, or or the landing made safely. And those tools build in a whole lot of margin already. They're only valid if you actually follow proper procedures though.

By your logic we should mandate 20,000 foot long runways for those for whom 10,000 feet is not enough.

[photofly]

You've completely missed the point. No amount of margin will suffice if crews make multiple gross errors. ...
By your logic we should mandate 20,000 foot long runways for those for whom 10,000 feet is not enough.

And by your logic, we shouldn't mandate any margin at all. Crews should simply not make mistakes, then margins are not required.

The point is that people do make mistakes, and those mistakes will kill people. To err is human. Margins acknowledge that, and save lives.

[complexintentions]

WTF are you talking about - where did I say that there should be no margins at all? There already ARE performance margins built into every aspect of takeoff/landing operations: factoring of landing distances and the like, time allowance for the recognition of a failure before initiating an abort, and so on. And of course there are overrun strips in place. No one expects perfection in human or machine performance, fact is the allowances are already pretty generous in this regard.

My point is that margins should be there to protect against minor errors, not gross ones that are entirely preventable. Like choosing to land high, fast, long, in a thunderstorm, and then not using proper braking technique. Adding even more "margin" to account for that type of event goes way beyond designing for the average.

I mean, if you can't properly calculate landing distance, properly brake the aircraft, or execute an aborted takeoff, perhaps it's just better if you don't get behind the controls?

I guess there are two schools of thought: one is to design things around the absolute lowest common denominator, the other is to expect performance to a pretty reasonable and easily-achievable level, with some safeguards built in for small imperfections.

I prefer the latter, but that's just me. I'm old-school like that.