I am taking responsibility for all the numbers and I can see/check all the data each time.
And that is the crux of the matter. djpil, apart from being a fine fellow, is both a very experienced and competent engineer and WCO and, without stating it in as many words, is taking steps to ensure that he can defend his actions before CASA or, far more importantly, in court, were a point to go that far for whatever reason.
Curious, what is the definition of a "Qualified person"?
I suggest, one who would be held to be an expert witness by a competent court of law (or hold the appropriate regulatory delegations or appointment as an authorised person by CASA) .. and that would include both djpil and me (although I think both of our CASA engineering appointments are now well lapsed).
Having come from a software engineering background and having written software which had airworthiness implications for defence, I'm wondering how this was prohibitive?
Likewise, I have had an extensive background playing with military stuff, albeit not in software design. The point of the discussion is who carries the PI for the activity ? The typical WCO is a one-man band and he/she needs to be darn sure that what is signed off will stand up to serious scrutiny in court. Been there, done that and very glad that I approach these things with a conservative philosophy. Those WCOs who might just sign off some unknown EFB software are, in my considered opinion, somewhere between very silly and very game.
Given the WCO would be using a a tried and tested set of formulas for each graph, so it wouldn't take much to produce a test dataset of every permutation and combination of values which could be verified against the base formula.
I don't have a problem with that comment for software developed by the WCO. However, if the WCO is going to develop the software, the problem, time and cost (as you would well know) lies in idiot proofing the software rather than the simple tech work. Again, although dated, I have been there and spent inordinate amounts of time so protecting software coding. Someone has to pay for that time and it isn't me. It might be fine for commercial software developers to hide behind the spurious claim "it's not a bug, it's a feature", but the PE and WCO can be hauled into court to defend his/her actions and work and is at risk of serious legal penalty .. think many dollars and, even, jail time. Hence my comment above -
those WCO who might just sign off some unknown compiled EFB software are, in my considered opinion, somewhere between very silly and very game.
so it wouldn't take much to produce
Have you done so for a complex loading system on a larger aircraft ? No ? I didn't think so. Think mucho time and effort, keeping in mind the aim is
nil defects, not just commercially acceptable output.
Dave, on the other hand, observes
I picked up my favourite EFB and clicked on the W&B for the type that I often fly and found many significant errors. Surprise, surprise. Now I know from anecdotal information, that there are WCOs who sign off on EFB or similar software without more than a cursory check. Brave folk in my view. It is so easy to cover all bases and put the responsibility back where it belongs .. on either the owner or operator. This can be done quite easily for EFB style programs. Simply get the WCO to review the program, develop a paper based image of the box output and go down the same path as for home grown software. Alternatively, pay for the time in thorough checking and we can do that as well. It remains the owner's or operator's choice.
How many accidents or incidents would have been avoided if W&B calculations had been done on an approved piece of paper rather than an unapproved electronic system?
Moot point in the absence of investigatory data.
If a pilot is willing to take the risk that W&B calculations done by an electronic system are not accurate, why would that same pilot bother to do accurate calculations using the approved paper system?
Touché and very pertinent to the discussion, I suggest. But, then again, such pilots fall into the class of under-performing pilots.
My experience is that it's pretty simple to confirm that calculations done by an EFB W&B system 'match' those done using the approved paper system, at the various extremes of the envelope.
I'm afraid your experience doesn't quite cut the mustard in respect of what is required to be done.
And on the paper system a pencil thickness can 'weigh' kilograms and accumulate to an inch of difference anyway...
Unfortunately, many loading systems of a graphical nature are poorly designed and your comment is valid. However, design the system well and the problem fades into background calculation noise