Question on Average Hours per Year

[wanttaja]

I'm trying to compute the average number of GA hours flown in the US for personal/recreational flying.

In 2020, the annual FAA GA Survey said that fixed-wing piston aircraft flew an average of 95 hours. But only 45% of the fixed-wing piston hours flown by GA are for personal/recreational use. An estimate of the annual flight hours flown for personal or recreational use would give a better comparison to Experimental Amateur-Built aircraft.

So: How to compute an average annual hours for GA aircraft operated for personal use?

Most classic aircraft are probably NOT flown for non-recreational use. Cessna 120s, 140s, and 170s, Bellancas (both the four-seaters and the Citabria-heritage), Aeroncas, J-3s, etc. How can we estimate how many hours per year these planes are flying?

One way occurred to me: Many aircraft suffer multiple accidents over their lives. For instance from 1998 to 2020, I found over 500 aircraft that suffered two (or more!) accidents. If the NTSB recorded the total time for the aircraft at the time of the accident (as they usually do), one could compute the average annual flight time for the period between the accidents.

I ran the analysis, and came up with what looks like reasonable results. Except for the outlyers.

Here's a summary of my data.

The middle column shows the median annual flight hours, the "Max" column indicates the maximum value within that aircraft type.

I'm just basically curious on what folks think is a "believable" maximum annual utilization rate for an owner-flown aircraft. I've looked at the NTSB reports for some of these, and it's obviously the same airplane. The later report for the high-scoring J-3 even referenced the previous report.

Ron Wanttaja

 

[Dana]

The higher numbers may reflect aircraft more likely to be used for training or cross country, increasing the yearly hours flown.

Me, I average around 50 in my Hatz.

 

[frfly172]

I do a minimum of 160 hours a year in my c150.

 

[Salty]

Statistics is a bit of black magic to me. Wouldn’t aircraft that fly more, statistically be more likely to have multiple accidents? Or is that begging the question? I guess folks that don’t fly much are more likely to have multiple accidents. My head hurts.

 

[GeorgeC]

55 +/- 20 sounds entirely plausible for us weekend warriors.

 

[somorris]

55 +/- 20 sounds entirely plausible for us weekend warriors.

I agree. Frfly172 flies his 150 160 hours or more a year, but most owners don't have the free time to be able to do that. We just sold our airplane, and my goal was an hour a week. With family obligations, weather, illness, etc., I missed some weeks but maybe could make it up the next week (or not). Also, annuals seemed to take several weeks, and we were down one year over six months for an engine overhaul. So, my recreational/personal flying wound up around 50 hours or so averaged over 20 years of ownership. Before I retired, I used our plane for some business, and usually got up to around 75 hours plus or minus. I rarely got to the minimum 100 hours/year needed to (as some insist) justify owning an airplane.

I think your analysis is pretty good, Ron, and likely as accurate as you will be able to get. Actual hours flying (especially after the new wears off) is probably is a lot less than some folks would say if some kind of proof is not required.

 

[RyanShort1]

Ron, should we start a voluntary effort??? I know my recreational flights were very few last year... even thought I probably spent well over 600 hours on board aircraft.

Also, as regards Cubs vs Lucombes... I suspect a rather higher number of Cubs are actually "working" vs Luscombes.

 

[wanttaja]

Ron, should we start a voluntary effort??? I know my recreational flights were very few last year... even thought I probably spent well over 600 hours on board aircraft.

Certainly is one approach, but I prefer to use processes that are reproducible by others. Could do a survey here, but others might complain that I hadn't engaged the Bonanza people on Beechtalk, or any number of other specific forums. Using the NTSB accident record means that anyone can repeat it, leaving just the methodology as a potential issue.

Also, as regards Cubs vs Lucombes... I suspect a rather higher number of Cubs are actually "working" vs Luscombes.

Yet another pet peeve of mine. The FAA GA Survey provides an estimate of how many of the GA fleet are active....76% of all GA aircraft, but only 63% of the homebuilts. We have no real split as to aircraft types. So any analysis has to assume that the same percentage of Cubs are active as are Luscombes, and the same percentage of Fly Babies are active vs. RV-10s. There *really* isn't any granularity available, which is why I try these end-arounds with the data.

The FAA Survey *does* estimate that 62% of the US aircraft with 1-3 seats are active; since most homebuilts have one or two seats, this pretty much corresponds. Though the FAA estimates that there are about 50,000 aircraft on the registry with 1-3 seats, and about half of those are probably homebuilts.

Ron Wanttaja

 

[GeorgeC]

ADS-B could give a pretty good answer for some airspace, but I don't know if one could extrapolate it to the whole fleet.

 

[texasclouds]

87% of all statistics are made up.

 

[Dana]

The FAA GA Survey provides an estimate of how many of the GA fleet are active....76% of all GA aircraft, but only 63% of the homebuilts.

I suspect that pilots of homebuilts, likely being more independent than the average Cessna driver, are less likely to fill out the survey, less likely to report accidents, etc., further skewing the statistics.

 

[wanttaja]

I suspect that pilots of homebuilts, likely being more independent than the average Cessna driver, are less likely to fill out the survey, less likely to report accidents, etc., further skewing the statistics.

I don't disagree, but: The survey results are based on the responses *received*. If a homebuilt owner doesn't reply, it doesn't affect the results of the GA survey. This led to the survey claiming a ~4000-plane drop in the number of ACTIVE homebuilts during the period when the FAA started the re-registration process (2010-2013).

Accidents, yes...homebuilt owners are more capable of hiding accidents than owners of production aircraft. The aircraft are often smaller and we have better capability to disassemble and haul the aircraft from a crash site. I know of two cases just in my EAA chapter.

Ron Wanttaja

 

[Lindberg]

I'm trying to compute the average number of GA hours flown in the US for personal/recreational flying.

In 2020, the annual FAA GA Survey said that fixed-wing piston aircraft flew an average of 95 hours. But only 45% of the fixed-wing piston hours flown by GA are for personal/recreational use. An estimate of the annual flight hours flown for personal or recreational use would give a better comparison to Experimental Amateur-Built aircraft.

So: How to compute an average annual hours for GA aircraft operated for personal use?

Most classic aircraft are probably NOT flown for non-recreational use. Cessna 120s, 140s, and 170s, Bellancas (both the four-seaters and the Citabria-heritage), Aeroncas, J-3s, etc. How can we estimate how many hours per year these planes are flying?

One way occurred to me: Many aircraft suffer multiple accidents over their lives. For instance from 1998 to 2020, I found over 500 aircraft that suffered two (or more!) accidents. If the NTSB recorded the total time for the aircraft at the time of the accident (as they usually do), one could compute the average annual flight time for the period between the accidents.

I ran the analysis, and came up with what looks like reasonable results. Except for the outlyers.

Here's a summary of my data.
View attachment 111304
The middle column shows the median annual flight hours, the "Max" column indicates the maximum value within that aircraft type.

I'm just basically curious on what folks think is a "believable" maximum annual utilization rate for an owner-flown aircraft. I've looked at the NTSB reports for some of these, and it's obviously the same airplane. The later report for the high-scoring J-3 even referenced the previous report.

I think there's probably a significant selection bias in this data. And the answer likely varies widely with type. I haven't read the MyFlightbook privacy policy, but I wonder if @EricBe could generate some anonymized stats on this.

 

[RyanShort1]

Certainly is one approach, but I prefer to use processes that are reproducible by others. Could do a survey here, but others might complain that I hadn't engaged the Bonanza people on Beechtalk, or any number of other specific forums. Using the NTSB accident record means that anyone can repeat it, leaving just the methodology as a potential issue.

Or it could be a simple website survey where any respondents could enter info and it could be shared widely across forums, etc…

 

[Tantalum]

I discourage both average and median as they don't describe the sample well. Nor should they be interpreted as causal. People however often misuse this to drive a certain point or story that tailor's to the authors wishes

A visual distribution represents the data much better, or even reporting on mode

Take this sample set
25
40
40
50
75
250
350

How much is the average person flying in hours per year? It would be fair to say that most people fly 40 hours per year.. it would be misleading to say the average person flies 120 hours per year even though that's what the average is. Most are flying well below that

$0.02

 

[wanttaja]

I think there's probably a significant selection bias in this data. And the answer likely varies widely with type.

Oh, certainly. No "modern" and only one "high performance" aircraft on my list.

But that's a necessary limitation forced by trying to restrict my list to those types that are mostly flown for personal and recreation purposes. This chart from the 2020 FAA Survey illustrates the issue:

Note the gold bar... the "Instructional" tally. Nearly 40% for the overall fixed wing piston, vs. less than 5% for the Experimental (and the Experimental listings include all sub-categories, PLUS Special Light Sport).

The aircraft flown for instructional purposes receive different maintenance attention, and the instructors themselves all have Commercial licenses. Does it reflect the portion of the certified GA fleet that are flown for pleasure? I don't think it does, really. Yet the homebuilt accident rate keeps getting compared to this commercial-centric baseline.

So...I started looking at common GA aircraft that probably AREN'T used for commercial or business purposes. Cessna 120s, 140s, and 170s, for example. Probably few of those on the rental/charter line, compared to 150s, 152s, and 172s.

The Bellanca Cruisemaster/Viking set is a good example. According to the FAA GA Survey, four-seat aircraft fly an average of 103 hours per year. But if the owner of a Bellanca sells it and buys an RV-10, he now only flies 45 hours a year....according to the Survey. If both types of aircraft suffer the same number of accidents in a year, the accident rate for the RV-10 will be double that of the Bellanca.

Using my method, though...the RV-10s seem to accrue an average of 78 hours per year, vs. 84 for the Bellancas. To me, this is a lot more believable result.

(Note the averages are computed differently. The Bellanca average is based on the time and accrued hows between pairs of accidents to the same aircraft, while the RV-10 results are based on the homebuilt's total time at the time of the accident and the year of completion. And yes, I am aware of the potential problems when using two different systems.)

Had a few surprises along the way. Have Champion/Aeronca/Bellanca 7X and 8X series (Citabrias and Decathalons) in the mix. Their annual hours came out significantly higher than most of the production airplanes in my set. I believe that reflects the Decathalon's use as an aerobatic trainer; much of the operations might well be in rented airplanes vs. owner-operated ones.

The Piper J-3 totals seemed to come out a bit high, too. The two high-utilization cases cases I looked at were apparently commercial operations...one rented by an FBO, the other one having two accidents (two years apart) involving students losing control (e.g., also probably a commercial operation).

So, where am I taking this? Like 90% of my accident analyses, it'll probably end up being just to satisfy my own curiosity. The methodologies for computing the average annual flight time for both the production set and the homebuilts are kinda loosey-goosey. Not sure if I'm willing to commit to them in print.

Ron Wanttaja

 

[wanttaja]

I discourage both average and median as they don't describe the sample well. Nor should they be interpreted as causal. People however often misuse this to drive a certain point or story that tailor's to the authors wishes

A visual distribution represents the data much better, or even reporting on mode

Take this sample set
25
40
40
50
75
250
350

How much is the average person flying in hours per year? It would be fair to say that most people fly 40 hours per year.. it would be misleading to say the average person flies 120 hours per year even though that's what the average is. Most are flying well below that

$0.02

But how does this argue against using the median?

The average of that set is 118 hours, which I agree isn't really reflective of the actual experience level. But the median is 50, which is must closer to what we would consider "typical" in such a set.

Ron Wanttaja

 

[RyanShort1]

I think that Citabrias and Decathlons get used a whole lot more for training and there are places like LeTourneau University that use them commercially, and I even know of a pipeline company flying some.

 

[EricBe]

I think there's probably a significant selection bias in this data. And the answer likely varies widely with type. I haven't read the MyFlightbook privacy policy, but I wonder if @EricBe could generate some anonymized stats on this.

I'm not sure I can do a whole lot better than the graph above - I can only infer flying type based on aircraft type (Boeing 787 is probably not GA, C172 probably is). But here's a quick and dirty query swag based on flights on MyFlightbook since Jan 1 2022 - looks like it's pretty skewed towards fixed wing and helicopter - the other stuff - which are decidedly non-zero - don't even show up on the chart because they're just so swamped. I'm defining instruction here as any flight that has EITHER dual received OR instruction given. I think that what you're seeing - no surprise - is that the multi-engine jets are being used for business, and the single-engine pistons are being used for training. No surprise there. Note that there can be some double counting of instruction if both the instructor and the student log the same flight, but I doubt it's enough to skew the shape much.

 

[sarangan]

I'm trying to compute the average number of GA hours flown in the US for personal/recreational flying.

In 2020, the annual FAA GA Survey said that fixed-wing piston aircraft flew an average of 95 hours. But only 45% of the fixed-wing piston hours flown by GA are for personal/recreational use. An estimate of the annual flight hours flown for personal or recreational use would give a better comparison to Experimental Amateur-Built aircraft.

So: How to compute an average annual hours for GA aircraft operated for personal use?

Most classic aircraft are probably NOT flown for non-recreational use. Cessna 120s, 140s, and 170s, Bellancas (both the four-seaters and the Citabria-heritage), Aeroncas, J-3s, etc. How can we estimate how many hours per year these planes are flying?

One way occurred to me: Many aircraft suffer multiple accidents over their lives. For instance from 1998 to 2020, I found over 500 aircraft that suffered two (or more!) accidents. If the NTSB recorded the total time for the aircraft at the time of the accident (as they usually do), one could compute the average annual flight time for the period between the accidents.

I ran the analysis, and came up with what looks like reasonable results. Except for the outlyers.

Here's a summary of my data.
View attachment 111304
The middle column shows the median annual flight hours, the "Max" column indicates the maximum value within that aircraft type.

I'm just basically curious on what folks think is a "believable" maximum annual utilization rate for an owner-flown aircraft. I've looked at the NTSB reports for some of these, and it's obviously the same airplane. The later report for the high-scoring J-3 even referenced the previous report.

Ron Wanttaja

Flight training and recreational flying would be very difficult to separate. Some may fall in both categories. I am not so sure if classic airplanes constitute the majority of personal flying. Of all the pilots I know who fly personal trips, only a tiny fraction use classic airplanes.