Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
  • #33351


    My objective is to measure “efficiency” of a process from cost/resource/process…Problem is it is a really big process that runs across the company, multiple countries, multiple sub-processes..
    Has any one done something like this? Tips/suggestions to approach appreciated. thank you



    Measure the efficiency for each subprocess across the company and then multiply these efficiencies… like we do for a rolled throughput yeild…. what u get as a result of multiplication will be your overall efficiency.



    Begin with a Value Stream map of the entire process, include cycle times and value added and non-value added steps.
    Once completed analyze the process



    Your huge global process sounds like its too big to tackle all at once.  I like the idea of a top level value stream map and CTQs.  You need to scope this huge process into manageable chunks.  This may possibly be a dozen Six Sigma projects.
    No problem is too huge for Six Sigma when executed effectively.  Good luck.


    John Hargreaves

    Yes, I have worked on global processes and it’s not easy, especially when different businesses and national cultures and languages are involved. 
    Although I agree with the need for practicality and therefore a ‘scoping down’ or a ‘chunking’ to sizeable bits that the other contributors have suggested, you need to be careful not to lose the connectivity of the overall process at hand.  The art is in knowing how the separation into ‘bite-size’ projects (process component bits/pieces) can give you a solution that is more than the some of its parts and not the opposite, i.e. an overall sub-optimization.  A lesser ‘bit’ might make for a larger ‘whole.’
    Processes tie people and technology together, the technology being all forms of hardware (not just computer h/ware but all supplier materials) and software inputs.  Added to this are the myriad business rules and policies that can speed up one process yet slow down another, often unknowingly.
    A total systems thinking approach is appropriate here, one that acknowledges the dangers in process separation and attempts – like we do with functional business units – not to allow a stove-piped approach to problem resolution result.
    My thoughts are that the solution is to (1) acknowledge the danger in ‘separation’ – by all team members on all process bits, realizing that sub-optimization at the process ‘bit’ level can be desirable if it translates into optimization at the ‘total’ process level (2) determine how to communicate potential adverse impacts – one process on another – through changes/improvements that will occur as improvements are put forward (3) test all potential impacts as they are discovered/realized (4) teamwork at the ‘bit’ and ‘total’ process level – the ‘total’ process level team being composed of representative members of each ‘bit’ team.
    Although it may seem logical to do all this work together, reality is that it will be very difficult to co-ordinate.  My suggestion is that you first identify and prioritize what parts are driving most of the problems and tackle these first.  Notwithstanding this, you must endeavor to keep the ‘connectivity’ issue described above in mind and test whatever may seem to be a cure (solution) at the process ‘bit’ level to determine that it is not worse than the sickness (problem) at the whole process level.



    You’re describing the E2E (End-to-End) approach. The best
    way to measure efficiency is to 1) undestand/validate the
    primary, business and financial metrics of this E2E process; 2)
    map out the E2E process; 3) identify key failure points
    throughout the process taken into consideration all the CTC’s
    and CTQ’s; 4) identify all the non-value added activities,
    waste, redundancy & complexity steps throughout the
    process; 5) validate financially the waste/non-value added
    activities and the complexity of the process, if possible; 5)
    identify DPMO’s and sigma levels in each step of the process;
    6) identify entitlement of the process; 7) finally measure
    baseline vs. entitlement and metrics. This is not easy. Rather it’s a long process. Usually in large
    corporations, this process takes ~ 2 years. What’s your



    try using a dpmo method for each subprocess….if it streches across multiple locations do a dpmo first by location for that subprocess and then add up all the dpmo’s to obtain a process efficiency ….or u cud try using average dpmo method wherein u cud average out the dpmo to obtain the process level sigma…….

Viewing 7 posts - 1 through 7 (of 7 total)

The forum ‘General’ is closed to new topics and replies.