What does "reliability" mean to you?

Reliable to me means minimal failures outside of standard maintenance.

When searching for a reliable vehicle I am typical interested in low maintenance as well.  I agree with bob's remarks regarding what acceptable maintenance intervals are.

Not having to fetch the wife in her 50yo VW that runs fine...

It's been having old car issues and she was in no mood.

Driven5 said:

If car A requires more maintenance than car B to prevent failure, then car B is still more reliable than car A even if both never break down.

Both are equally reliable, but car B has lower maintenance costs.

codrus (Forum Supporter) said:

Driven5 said:

If car A requires more maintenance than car B to prevent failure, then car B is still more reliable than car A even if both never break down.

Both are equally reliable, but car B has lower maintenance costs.

Car A causes more disruption than car B. More disruption is less reliable. Otherwise any car could be made 'equally reliable' simply by sufficiently increasing both the frequency and scale of maintenance required.

Cost of the maintenance is unknown here.

I laugh when people say they're car is reliable (usually VW owners) and then tell me about all the things they've had to do to it. Change the DSG fluid, fix some other cam thing, blah blah blah. But it's A888 is "bulletproof" they tell me! We don't have shared reality around reliable. 

I like Bobzilla's definition.

To me bulletproof describes the ability to take abuse... not reliability. 
 

VW can be very reliable with proper maintenance, but I think you will often find large discrepancies between what the manufacturer recommends and what enthusiasts recommend.

Driven5 said:

codrus (Forum Supporter) said:

Driven5 said:

If car A requires more maintenance than car B to prevent failure, then car B is still more reliable than car A even if both never break down.

Both are equally reliable, but car B has lower maintenance costs.

Car A causes more disruption than car B. More disruption is less reliable. Otherwise any car could be made 'equally reliable' simply by sufficiently increasing both the frequency and scale of maintenance required.

Cost of the maintenance is unknown here.

See my earlier post re. "availability" versus "reliability".

In reply to AngryCorvair (Forum Supporter) :

Whenever something is not available,  it's not performing its intended function. Whether that is due to maintenance or repair is irrelevant, beyond a given amount of down time being less disruptive planned than unplanned. Availability is a component of reliability.

In reply to Driven5 :

This is really testing my memory banks, but I think there's a distinction in reliability between performing its function vs performing its function when asked. When something is unavailable due to planned maintenance, you're not going to ask it to perform its function, therefore it's unavailability does not affect its reliability.

but I graduated 32 years ago and what I've said here could all be 100% PFA.

Driven5 said:

Whenever something is not available,  it's not performing its intended function. Whether that is due to maintenance or repair is irrelevant, beyond a given amount of down time being less disruptive planned than unplanned. Availability is a component of reliability.

The concepts are best kept separate because that allows one to speak about them more precisely.

Would you consider a Boeing 747 to be "unreliable"?  Airliners need something like 5-10 hours of maintenance for every hour of flight, which dwarfs the maintenance requirements for any car out there.  Military aircraft are worse than that, 30-40 to one is not unusual.

Think about the related-but-not-the-same concepts of precision vs accuracy as it pertains to firearms, or fast vs quick in drag racing.  Blurring too many concepts together makes things confusing.

High reliability should not be confounded with low maintenance. They can go together but don't have to. 

Reliable is it does what it needs to when it needs to. We have an older Odyssey, with a lot of miles, and with a higher mile vehicle lots of things go sideways. What's amazing is that it rarely is something catastrophic where it can't be used. All sorts of things have worn out: door actuators, control and trailing arm bushings, wheel bearings, ball joints, brakes, tires, ac compressor, PS pump and hoses, alternator, battery, leaking valve cover gaskets, my latest evap issue (was the fn vacuum valve in the canister!), egr valve, drivers side window switches, two timing belts and waterpumps... that's just off the top of my head. Reliable? Definitely. Low maintenance? For almost 300k worth of use, maybe.

Why must one perform "maintenance" other than that the part being "maintained" is no longer useful for its job? AKA the maintenance item is either broken or soon will be. There is no difference between taking the time to rectify a maintenance item (predicted to fail) or a broken item (has failed not on a schedule). Therefore I consider "maintenance" to be the same as "fixing a breakdown". Just because the factory called it out in advance as a thing that will break/stop working effectively doesn't make it any less a thing that failed.

The fewer times I have to lift the hood, the more reliable it is. The less it asks of me in order to keep driving, the more "reliable" I consider it to be. 

In reply to P3PPY :

I disagree. Nothing is completely infallible, and certain parts of your vehicle will wear out due to the nature of their function. If your car needs brake pads, does that mean it's broken? No. It means the pads have reached the end of their useful service life. Same with spark plugs, brake shoes, tires, windshield wiper blades, and all the other wear items your vehicle uses. 

I think we're all pointing at essentially the same thing. Keyword: SURPRISE. An unexpected failure is a breach of reliability (trust), especially if it happens on the road. The relative proportion of maintenance a given machine requires in order to achieve a (presumed) level of surprise-proofness is an equation of wildly varying proportions each of us defines differently. Since reliability≈functionality-as-planned, I'll argue that ratio/tolerance also has a large bearing on vehicle selection overall. Some buy and run late-models with the expectation that the scheduled items in the owner's manual are all they require; there's nothing wrong with that. Some of us have never owned something that didn't get a complete powertrain and running gear replacement as a prerequisite to joining the fleet. There's nothing wrong with that approach either, but the level of effort is certainly greater. Personally, I'm in the second camp.

Machinery will wear out with use, but will wear out a helluva lot faster with disuse or misuse. The various bits that make up a car also wear at different rates, and I believe abuse increases that scatter. Some manufacturers are better at synchronizing component lifespan than others. I'd much rather do an engine-out replace-all job than keep up with a leaky old turd that "surprised" me every other week. I relish the certainty of a total rebuild I did myself, carefully checking and eliminating the potential for surprises along the way. Most of my stuff is old, with the newest being old enough to drink. Every one of them is stone-cold reliable but it sure took a lot of hours in the shop to get there.

There's a notable sub-set of universal components common to every car that I'd consider exempt from judgment. These are things like tires and batteries, things which are brand-agnostic. If your battery pulls an Elvis or your tire finds something pointy in the road, that's not related at all to the car's reliability, it's just a E36M3 day.

A story as told to me by an acquaintance:

The head of Maico asked him why do Americans consider our motorcycles unreliable?

Answer: because you can't race them hose them off and race them again.

Reply: these are race bikes, and they need proper maintenance. We list everything that needs to be done after a race.

Answer: the Japanese bikes don't need this.

Why do I bring this up? Because this is the standard of reliability I grew up with. 

If you have to "service" something every 40-50K that's a band aid for an unreliable car and or poor design.

I know some manufacturers would like us to believe that the frequent intervals justify the performance but it doesn't. There are plenty of cars with equal performance that don't need as much maintenance.  

If a car needs anything other than oil changes before 100K I consider it unreliable.

In reply to Recon1342 :

I wonder if you may be reading into "broken" something I don't mean, I'm not implying it failed in an unexpected way. Feel free to call it "reached the end of its useful life," it's the same thing as "broken," really. Either way, car no work right if no replace, correct? 

Your examples of brakes, wipers, etc. are just the same: if you don't replace then car doesn't work right.

You brought up routine maintenance items. Under my definition I'm not saying no car goes without breaking things/having things reach the end of their usable life. I'm just saying that the more "maintenance needs" a car has, the less reliable it is, because it has more thing that are broken/end of usable life AND THEREFORE interrupt my use of the car. 
 

Lets go to the extreme:

What's an example of an unreliable car? Let's say the stereotypical Subaru, realizing that not everyone will agree, but run with me. 
They blow head gaskets, right? A lot. And that's a real hassle to fix, esp compared to a wiper blade, right?

So what if Subaru listed them on their regular service schedule as a routine, planned maintenance item? Does that suddenly make them "reliable" just as long as you do the "scheduled maintenance?"

I like old cars!  They were made with easily serviced bits that might fail on occasion but were often fixable even at the side if the road.  Today they do seem to have longer intervals between issues but when they do stop on you, it is often because some 'module' failed and you either can't get at it or the manufacturer stopped production of replacements a year or two earlier.  Regular point ignition works well; one guy with a brand new Pontiac Solstice had an electronic module fail - it was during the aftermath of the Japanese tidal wave in 2011 and GM actually had to buy back his almost new car because they had no parts to fix it and didn't know if and when they would restart production.

A question for the people saying reaching the end of a components useful life and requiring maintenance is not different than a failure:

Tires wear out and need to be replaced, correct? Does that make them unreliable? 
 

This is quite the interesting discussion.  

Please continue.  

In reply to P3PPY :

In that extreme example, I'd say as long as the number of failures that occurred between change intervals was low enough (within total volume sold), then it would be considered reliable.

I would expect it to negatively impact sales, but that's a trade off the manufacturer needs to make.

They make those types of trade offs on all aspects of the design, using knowledge of what consumers are willing to spend to maintain the vehicle, impact of a failure, and acceptable use life. 

You could make brake pads last longer by making them thicker (and add cost and weight), increase oil change intervals by increasing sump capacity (increasing the cost per change), or use timing chains in place of timing belts (eliminate change intervals, but increase nvh and complexity).

Sure the maintenance could be less, but there's no free lunch and that maintenance cost would end up going into the purchase price so drivers would be paying it upfront instead of over time. 

No Time said:

A question for the people saying reaching the end of a components useful life and requiring maintenance is not different than a failure:

Tires wear out and need to be replaced, correct? Does that make them unreliable? 
 

That's a good question. And it hints that there may be gray area, so let's ponder it. 

So if you don't replace them when they're worn, you get what's called a "tire failure." And you can buy tires with longer or shorter usable life. So tell me, if you bought tires advertised as 60k but only got 40k out of them, would you not put them in the camp of other things you consider "unreliable?"

And of course one might argue that someone would buy a softer tire for better grip, but even then you have a benchmark of expectation for longevity given the particulars of the application.

So I guess while calling a tire "unreliable" isn't my normal vernacular for them, yeah, I'd say that if they wear out faster than normal/expected, then they're not something I can rely on.

And, as it pertains to the parent car that the tires are stuck on, we all know that the heavier the car, the more you will wear, so the car does play a part in the lifespan or "reliability" of that tire. 
 

HOWEVER, as it pertains to the wider discussion, semantics aside, given that all cars require wiper and tire replacement, I'd say that all cars share those maintenance/breakage points in common and therefore cancel one another out. That is, unless the mean time between failure/wear out/breakage was exceptionally short, in which case do you not suppose that one would begin to complain about how "such and such a car just eats up brakes/tires/etc.," which I would say there's therefore an argument there that the car could fall under the "unreliable" umbrella. 
 

Could be a stretch, but maybe not? Thoughts?

No Time said:

A question for the people saying reaching the end of a components useful life and requiring maintenance is not different than a failure:

Tires wear out and need to be replaced, correct? Does that make them unreliable? 
 

If we are talking like for like; If I have 300-400 tread wear tires and one sets useful life is 80K and the others are toast in 30K, then yes the 30K tires are unreliable because if I push them past 30K they will fail.

The point for me is like for like performance. If there are other manufacturers offering equal product that lasts significantly longer, than I consider the one that wears out sooner unreliable.

Reliability is not binary, it's a spectrum.

Most of the debate here really seems to be about operational definitions vs technical definitions.

In reply to P3PPY :

I see what you meant now, P3PPY. A known issue that crops up multiple times with a certain make/model usually indicates an issue. If the vehicle "just eats" tires, brakes, head gaskets, etc... then yes, it is proving itself unreliable. 

This was going to be an edit but I'm wordy so it's an add-on. 
 

If you design a car with a wearable item that many other manufacturers do not have, then that seems unreliable to me. Timing chains that No Time brought up are a pet peeve of mine. If Honda recommends that I spend $1500 every hundred thousand miles but a Chevy Cavalier doesn't, then that's a score against the Honda because the stupid ENGINE will break if the serpentine belt THAT I CANNOT SEE wears out.

To me, and I'm beating a dead horse here, "scheduled maintenance" is just a mfg's marketing way of making their known point of failure palatable.

 It doesn't in toto make the Chevy more reliable, of course.

Take this ten years into the future:

Might a prospective car buyer consider an EV more reliable than ICE because of the latter's regular habit of wanting oil, transmission fluid, belts, spark plugs, etc., and more regular brake changes? I think they might. And I wouldn't fault them. Because without that "routine" time in the shop, the car doesn't run.