MIT Mid Air Collision Paper

[oregonboy109]

I want to share a paper done by MIT on mid air collisions.

I was suprised that a large percentage of the collisions involved aircraft on the same heading...

pretty interesting stuff.


View attachment mit.kunzi.midairs.pdf

 

[ahkahn]

I guess it makes sense to me... Either low-wing/high-wing on base/final or parallel aircraft converging slowly on the sides and cannot see each other.

 

[dtuuri]

I think the non-formation cruise chart is the best representation of risk. There, you can see that when only one set of eyeballs are scanning ahead (from behind the target) the risk increases as the closing rate increases until, at about two o'clock, the target's eyeballs begin to contribute to collision avoidance also. From that point the risk drops dramatically, but then climbs again as closing speed continues to increase to the twelve o'clock position.

So... "see and avoid" by looking everywhere blocked by the plane's structure, and avoid high closing speeds seems to be the lesson when approaching an airport.

dtuuri

 

[acrophile]

So... "see and avoid" by looking everywhere blocked by the plane's structure, and avoid high closing speeds seems to be the lesson when approaching an airport.

On the other hand, high closing speed increases your glide range in the event of engine failure (you still want to pull back to best-glide speed, but starting at a higher speed will boost your range). So high closing speed compensates for the danger of being at low altitude as you approach the pattern (depending in part on the surrounding terrain).

I routinely perform shallow S-turns as I approach an airport. This practice lets me see more of my surroundings, probably makes me more visible to others, and above all makes it far less likely that my approach is aligned with another plane's, leading to convergence.

It may even be a good idea to conduct shallow S-turns throughout the pattern itself for the same reasons, though I haven't been doing that.

 

[dtuuri]

On the other hand, high closing speed increases your glide range in the event of engine failure (you still want to pull back to best-glide speed, but starting at a higher speed will boost your range). So high closing speed compensates for the danger of being at low altitude as you approach the pattern (depending in part on the surrounding terrain).

I routinely perform shallow S-turns as I approach an airport. This practice lets me see more of my surroundings, probably makes me more visible to others, and above all makes it far less likely that my approach is aligned with another plane's, leading to convergence.

It may even be a good idea to conduct shallow S-turns throughout the pattern itself for the same reasons, though I haven't been doing that.

The wing waving improves your visibility to others, provided they're looking your way. I don't buy the speed value in a risk/benefit analysis, though. It might salve your neurosis against an engine failure at the cost of a midair for your converging target.

dtuuri

 

[steingar]

The number of midair collisions yearly is nearly almost always vanishingly small. Far more pilots perish from flying into IMC or running out of gas.

 

[Checkout_my_Six]

....and a good many mid-airs involve ATC.

 

[acrophile]

The wing waving improves your visibility to others, provided they're looking your way. I don't buy the speed value in a risk/benefit analysis, though. It might salve your neurosis against an engine failure at the cost of a midair for your converging target.

Engine failures are more common than mid-air collisions, so it makes sense to take precautions for the former--especially in combination with shallow S-turns, which make it almost impossible for two planes' flight tracks to accidentally remain the same as they converge.

 

[PaulS]

Although most of the comments here are true I would posit that one who is not vigilant or worse has their head buried in the cockpit astronomically increases their odds of a midair.

 

[dtuuri]

Engine failures are more common than mid-air collisions...

What do you mean "more common"? I bet that if you calculate the engine failure rate per operating minute it's far less than the pattern midair collision rate per pattern minute.

dtuuri

 

[coloradobluesky]

Just about the only place mid airs occur is within 5 miles of the airport. All but one of my close calls were close in. Not sure if the MIT reports that, but thats what I have learned.

I suppose we all know that airplanes are easier to see from lower altitude. Looking down they get lost in the clutter.

Just my 2 cents.

 

[John Collins]

Just about the only place mid airs occur is within 5 miles of the airport. All but one of my close calls were close in. Not sure if the MIT reports that, but thats what I have learned.

I suppose we all know that airplanes are easier to see from lower altitude. Looking down they get lost in the clutter.

Just my 2 cents.

During the day they get lost in the bright sky. The eye only detects them by the contrast of dark verses the bright sky background. Landing lights make this effect worse.

 

[acrophile]

What do you mean "more common"? I bet that if you calculate the engine failure rate per operating minute it's far less than the pattern midair collision rate per pattern minute.

Good point. I don't have data on collision rates in the pattern--or, more pertinently, during approach to the pattern (since I do slow down before the pattern itself). But making some reasonable assumptions from published data, those two rates seem roughly comparable. So I retract my claim that the engine-failure rate is higher.

I'd still guess, though, that mid-air collisions approaching the pattern become vanishingly unlikely with the added precaution of S-turns. But I admit I don't have data to back that guess up.

 

[JohnAJohnson]

For collisions away from the airport, the takeaway is that most happen with both at cruise. I think I'll start flying 30' either side of the 500' altitude.