(07-03-2016, 10:11 AM)dorothy.pipet Wrote: Another attempt for comments please.
I feel I have taken a different approach from the others on here so some brief indication how successful it is would help.
I didn't think that this was as good an answer as you usually give. As ever it is of right length, clearly presented etc. The thing that made me feel it had fallen short was that the second two pages were not enough tied into the particulars of the first page. It did get better towards the end but it certainly started off too abstract and could be seen as regurgitation of Yellow Book. To be fair the question paper does date from 2005 and that sort of answer might have been more acceptable to the examiners in the context of that time, but that is not what they are looking for nowadays and possibly word the questions a bit differently to nudge the candidate to give the form of answer they want.
Considering the risks that a manufacturer faces, I think that I'd have considered various categories-
a) Financial outlay in getting to market initially [your item 1 slightly reworded and part of your item 5]
b) The uncertainty regarding for which project it will actually be ready and therefore a potential damages claim and reputational loss for being unable to achieve a project commissioning date,
c) Being able to sell it for a profitable price in adequate quantities to recoup that cost, so dependent upon things outside your control such as the client's funding to do suitable projects at all; rival products from competitors coming along which may be preferred for understandably good reasons (better functionality, cost) or perhaps due to a policy decision that may not be so obvious and may appear to be on a whim. [very much expanded second part of your item 5]
d) Having sold it and hence committed to support it for a period of years, then have the uncertainty of how can deliver this [your item 4 plus an element of your item 3 re maintaining the skill base in-house to continue to support]
e) Once it is in the field in perhaps some quantity, the chance that things may emerge which mean that some element of the RAMS performance turn out to be considerably lower than had been anticipated and therefore having reputational loss, costs of remedial, loss of future business etc.
Whereas I don't disagree with any of the items you listed, to me you did not always make it sufficiently clear what was THE LOSS which would be incurred in each case; hence the spin I placed on what I have written above.
I think that I may well have included your items 2 & 3 within my item a); some of the reasons that lead to protracted development timescales that eat up budget as well as push back dates include unclear / misunderstood specifications, false assumptions, lack of clarity over interfaces which all conspire to mean that the required task is underestimated and are often coupled with an over-confidence of what staff would be capable of doing. As written I felt your item 1 was too bland; if one persists for enough time, throws enough money at it, is prepared to start over time and time again then one will eventually get to a solution that does work, so the risk is not so much that it can't achieve as that it can't be achieved sensibly.
There are only 5 marks for these 5 risks but I think that you could have drawn out these various threads when moving into the second part of the question, by which I mean rather than putting all the detail on the first page including these various thoughts as sample entries in the Hazard Log. This would be the place to identify the risk that the designers did not appreciate enough about the environment or the interfaces needed and therefore would result in much re-work and evolution from the original concept.
I think this question needed a slightly less System Engineering detail and slightly more broad consideration of risk; for a supplier this is more focused on the financial business risk than anything else. Although the question was asked ten years ago it is interesting to look at in the context of 2016 in the UK:
# Siemens SIMIS-W customized for NR, 2 installations only
# Bombardier Ebilock, never made it into service
# Ansaldo SEI, 1 installation on the Cambrian ETCS, perhaps another is coming in a rather different implementation,
# Ansaldo ACC, 1 installation Manchester South,
As very large international companies, these experiences did not threaten their ongoing existence, but in each of the cases there were employees in the UK who lost their jobs- they may not have been involved in the interlocking development and implementation but the hoped for ongoing projects never materialised and they then proved surplus to requirements and often whole offices were closed down. Hence the various debacles did threaten future prosperity of individuals of their workforce. In such a scenario it is understandable that attention is not always fully on the task in hand, so there could then be a potential safety threat to the railway as this could be a causal factor to an accident.
Then think about Modular Signalling. So far two suppliers have delivered between them 3 modular schemes; the whole idea being "plain and simple, cost reduction by keeping things standard. How many versions of NR Modular Signalling handbooks have there been- 3!
Where is the string of schemes to bring the economies of scale?
Where is the discipline in the client for stopping all the"slightly bespoke" solutions here there and everywhere that move away from the vision?
It was originally about replacement of semaphore signaling with 2 aspect signals and perhaps the odd 3 aspect. Generally only a couple of routes from a signal, perhaps 3 needed occasionally. No banners. No permissive.
and where are we now...............?
Hence one of the big risks is to have a product tailored for one client.
Another is that client keeps changing it's ideas and the more that one has tried to template everything in order to deliver production line style to get the efficiencies of mass production, the more difficulty one faces when that original certainty comes apart at the seams........
