Engineering

Previous post showed F-35 obsolete

   The biggest mistake made in giving contracts for engineering , particularly military design, is to treat each contract as a single entity without any connection to anything that came before.  This is typically combined with awarding credit to the previous contractor within any realm of work for experience, no matter how badly mangled the previous work was done.
   The only way to see improvements in the quality of engineering is to promote project managers who do good work and remove those who do not from management positions.  This would involve some form of peer review, which can always lead to a buddy system of helping each other, but that is little more damaging than the current system, of bogus bidding.
   In the current system, one must be a qualified bidder.  But the qualifications often involve having done the work previously.  That makes it impossible for any new company to enter the bidding process.  The only way around it would be to hire someone who was a senior manager from a company that has done the work previously.  But if the previous company  were complete screw-ups, that would mean, most likely, hiring someone who was responsible, at least in part, for the screw-up., guaranteeing the progression of incompetence for another project cycle.
   Another way to be disqualified is to have a company that is too small.  One might protest that a competent company could do the work with a small workforce, but that would be futile.  This guarantees,that large, over-bloated, incompetent companies will always produce bad, over priced work on all contracts.
   The measures of competence in engineering are two-fold; having a working design at the end of the project cycle and the number of iterations it takes to produce the working design.  An iteration is a complete pass through the design, in fact a complete design, or design of some part of the complete project.  Three iterations would be world class engineering; a preliminary design, a design to correct conflicts and deficiencies and then a final design.  It is rarely achieved.  If a project takes twenty iterations either; the project involves doing work never done before, the client is a moron who keeps demanding changes in performance or the people managing the project are thumb-sucking retards.  This is much more often to occur.
   Improving the process requires money and time.  There would need to be a rating of engineers by written and aural tests.  this would require several iterations; a preliminary test and verbal screening with an intial ranking of engineers followed by additional rounds of written and aural tests to give a finer grained rating of engineers by knowledge and competence.  The reason for this is that one can only rate people who are below one's own knowledge, if someone knows more, it cannot be judged how much more with any great accuracy.  So, the first iteration would produce a rough rating of; knows less than me, knows about he same as me, knows more than me.  The better engineers would then have to rate each other repeating the proces two or three times before the rankings would be fairly accurate.  It would than be a good idea to go through the rating of all engineers one more time, to try to clean up the results.
   Once the rankings are done, it would be a good idea to try and psychologically screen out people who should never be placed in charge of other people.
   Once all that has been done the fun can begin.
   Take engineers who have enough knowledge, competence and experience and, picking some at random, give them a relatively small project to manage.  If they do well, give them a larger project to manage. If they do not do well, give them a second chance, or, if they were a total disaster, remove them from the manager list.by doing this repeatedly, a core of competent efficient managers will be found,.  By assigning the better managers to significant contracts, those contracts will be completed in less time and with better performance.  It is cumbersome and unwieldy to establish, but once operational it will save money.  It may also require rewriting anti-trust laws to allow records to be kept and contracts to be awarded based on performance with other clients.
   It would also be a good idea to have the engineering staff rate the managers on a scale of zero to ten and use that as part of the overall rating, since they are the people who dealt with the management every day and they know if ti was incompetent.  But that kind of rating requires at least fifteen people on the project.
    Lowest bid contracting is useless, the clients almost always change the specifications during the project which means that the bidding was based upon nothing.  It works buying pencils but accomplishes nothing with engineering.