Performance characterization Don Clausing ? Don Clausing 1998 16.881 Fig. 1 Failure modes ? Noises lead to failure modes (FM) ? One set of noise values leads to FM 1 ? Opposite set of noise values leads to FM 2 ? Simple problem solving chases the problem from FM 1 to FM 2 and back again, but does not avoid both FMs with the same set of design values – endless cycles of build/test/fix (B/T/F) ? Don Clausing 1998 16.881 Fig. 2 Failure modes occurrence ? Failure modes occur when the essential performance characteristic deviates excessively from its ideal value ? Some failure modes are simply values that deviate so far that output is not useful; e.g., voltage from power supply is too low ? Some failure modes are catastrophic; e.g., paper that arrives too soon jams in copier ? Don Clausing 1998 16.881 Fig. 3 Performance characteristic ? What is good performance characteristic to use when reducing the occurrence of failure modes? ? Don’t merely count occurrence of failure modes ? Can’t distinguish between following two cases ? Don Clausing 1998 16.881 Fig. 4 Case 1 – easy to fix Occurrence of failure mode FM 1 FM 2 ? Don Clausing 1998 16.881 Fig. 5 Case 2 – difficult to fix Occurrence of failure mode FM 1 FM 2 ? Don Clausing 1998 16.881 Fig. 6 Case 1 and Case 2 ? Both cases have same failure rate ? But situations are very different ? Counting failure rate is very weak approach to the reduction of failure rate ? Concentrate on ideal function – What is the system supposed to do? ? Then make system do it all of the time ? Don Clausing 1998 16.881 Fig. 7 The engineered system Noise Signal System Response Control factors ? Don Clausing 1998 16.881 Fig. 8 Ideal function Ideal function (response) RESPONSE SIGNAL, M ? Don Clausing 1998 16.881 Fig. 9 Actual response Ideal response Effect of noises RESPONSE M 1 SIGNAL M 2 ? Don Clausing 1998 16.881 Fig. 10 Keep performance close to ideal ? Identify ideal performance (function, response) ? Then make actual performance stay as close as possible to ideal ? Linear response is called “dynamic” – desired value for response depends on input value of signal ? Don Clausing 1998 16.881 Fig. 11 Examples of dynamic response ? Car turning radius ? Car stopping distance ? Copy quality ? Casting ? Electrical resistance What are the signals? ? Don Clausing 1998 16.881 Fig. 12 Case study – hitch ? Used to connect implements to tractor ? Transmits power from tractor to implement ? We can all see its function ? But what is a good engineering statement of its function? ? Don Clausing 1998 16.881 Fig. 13 Functions of hitch ? Provide mechanical interface with implement ? Provide proper vehicle performance ? Meet ISO dimensional requirements ? Protect people from moving parts ? Don Clausing 1998 16.881 Fig. 14 More detail on first function ? Provide adequate performance in working range – Proper attitude through working range – Provide adequate depth – Provide adequate lift capacity at breakout – Provide Draft Control ? Provide adequate performance in transport mode – Provide adequate height – Provide proper kick angle – Provide adequate lift capacity at transport ? Provide easy hookup and disconnect ? Provide easy linkage adjustments ? Don Clausing 1998 16.881 Fig. 15 Yes, but what is ideal function? ? Meet ISO dimensional requirements ? Protect people from moving parts are important generic requirements, but are not elements of the ideal function. ? Don Clausing 1998 16.881 Fig. 16 Functions of hitch ? Provide mechanical interface with implement ? Provide proper vehicle performance are related to ideal function. Candidate for ideal function: Transmit load ? Don Clausing 1998 16.881 Fig. 17 Forces on system H I T C H IMPLEMENT TRACTOR F T F F F D W H F R F S ? Don Clausing 1998 16.881 Fig. 18 Noises in the field HARD ROWS ? Change in earth impedance causes forces to change ? Which changes do we wish to minimize? ? Don Clausing 1998 16.881 Fig. 19 Keep what constant? ? Constant force? ? Constant depth of engagement into the soil? ? Constant power? ? Don Clausing 1998 16.881 Fig. 20 Candidate ideal function Actual depth Ideal function (response) SIGNAL, depth set by farmer ? Don Clausing 1998 16.881 Fig. 21 Determination of ideal function ? Identify the performance variations that we would like to go to zero ? The performance that remains when the undesirable variations are zero is the ideal performance ? In the hitch case further analysis, tests, and discussions with customers are needed to identify (verify) ideal function ? Don Clausing 1998 16.881 Fig. 22 Ideal function ? Want Ideal Response to Signal – usually straight-line function ? Definition is often not trivial ? In the absence of explicit definition the objective of improvement activities is unclear; success unlikely ? Don Clausing 1998 16.881 Fig. 23 Signal/noise ratio ? Measure of deviation from ideal performance ? Based on ratio of deviation from straight line divided by slope of straight line ? Many different types – depends on type of performance characteristic ? Larger values of SN ratio represent more robust performance ? Don Clausing 1998 16.881 Fig. 24 Summary ? Knowing ideal function is crucial for success – we have to know where we are trying to get to, or it is unlikely that we will get there in a reasonable time ? Requires detailed engineering analysis of conditions for customer satisfaction ? Don Clausing 1998 16.881 Fig. 25 End