Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Performance indicator doubts in 4.4.1

Discussion in 'ISO 9001:2015 - Quality Management Systems' started by Qualmx, Jul 18, 2019.

  1. Qualmx

    Qualmx Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    443
    Likes Received:
    58
    Trophy Points:
    27
    Location:
    Mexico
    Hi everybody

    In the monitoring activity of the performance of a process, I detected some
    issues which makes me think what is the adequate manner.

    This is my case, an example

    A production process, at defining a goal = max rejects per month = 4
    Here I could consider that is allowable to have maximum 4 rejects a month.
    If I have less than 4 (2 or 3 is ok) , if is higher , (5 or more)
    I need to take actions,
    But
    What should be the action to take in this case,
    To perform a CA because I´m having more than 4? , I mean what is the reason I´m exceeding
    the goal?
    I don´t think that instead , I have to analyze the resulted rejects, because that have been addressed into the process by mean of attending complaints, corrections or corrective actions.
    Additionally I would be repeating the analysis.

    Is it possible to monitor the process performance by only monitoring the quantity of rejects and
    doing something when quantity is higher?
    Because on the side of the nonconformities, they are addressed properly.

    Another example somewhat different is:
    Production rate , goal = 15000 pounds per month
    If I reach 14995, that is ok, but if I get 7500, what?? it deserves a CA, that in this case
    I would need a CA because I only got half of production rate, the question
    What will be the analysis focused at? why the result was not met?
    or to analyze into the process the complaints, nonconformity reports, rejects?

    Please give some feedback

    Thank you so much
     
  2. John C. Abnet

    John C. Abnet Well-Known Member

    Joined:
    May 23, 2017
    Messages:
    535
    Likes Received:
    369
    Trophy Points:
    62
    Location:
    Upper Midwest- USA

    Good day @Qualmx ;
    Don't overthink this. Your organizations sets goals. Is there a plan to achieve these goals? When the goal is " missed" is the activity within statistical control? If so, then the result is the best the system can expect. Is the "miss" caused by an outlier or outlying activity? If so, then perhaps a corrective action is in line to correct that assignable cause.

    Consider the goals as part of a system/program/patterns. Unless the system/program has broken down and/or there is a company wide pattern of goals not being met (lack of planning, resources, etc..etc..) then corrective action for every single "miss" will likely cause more harm than good and overwhelm your teams.

    Hope this helps.

    Be well.
     
    Qualmx likes this.
  3. RoxaneB

    RoxaneB Moderator Staff Member

    Joined:
    Jul 31, 2015
    Messages:
    892
    Likes Received:
    1,033
    Trophy Points:
    92
    Location:
    Ontario, Canada
    I've talked about this approach several times on here...set appropriate nonconformance "triggers". A nonconconformance is a nonconformance is a nonconformance. However, not all nonconformances are the same.

    With my previous organization, we were really impacted by unscheduled delays in the production line. Not online did this impact our production numbers, but to start up after a delay could result in nonconforming product, so delays were a serious issue with us. So, we tracked the unscheduled delays, even if they were only 30 seconds all the way up to the ones that us delayed for much of the day. It goes without saying that the 30 second delays were not considered as risky as the longer delays.

    So, we reported all delays (and their cumulative amount of time), but we focused on the big ones...to start, let's say it was all delays greater than 2 hours. For those big delays, we did root cause analysis and tried to understand WHY the longer unscheduled delay occurred. The smaller delays were annoying (like mosquitoes), but they were not our focus. After a year of focusing on the >2 hour delays, we analyzed our unscheduled delay results and lowered our trigger to 1.5 hours. Now, in our 2nd year, any unscheduled delays greater than 1.5 hours would have that in-depth analysis.. And so on. With this level of analysis, we did notice some pockets or clusters of smaller unscheduled delays occurring in certain areas or attributed to certain activities - we turned these into projects for the appropriate teams, so, in a sense, those clustered, smaller delays became metrics with visibility to leadership.

    The idea is to focus on the larger, higher impact/risk issues than on ALL issues.

    Setting a target of 4 with no focus on the drivers behind the nonconformances is setting the organization up for a lot of paperwork, a lot of activity, and no unified path towards fixing the real issues. I suggest with continuing to track all nonconformances, but the actual KPI is the nonconformances associated with a specific root cause or problematic area/process.
     
    tony s, Qualmx and John C. Abnet like this.
  4. tony s

    tony s Well-Known Member

    Joined:
    Sep 10, 2015
    Messages:
    1,222
    Likes Received:
    913
    Trophy Points:
    112
    Location:
    Laguna Philippines
    You will notice from @RoxaneB comments, the word corrective action, even correction, was not mentioned. Do what your organization need to do to address the perceived problem. Know why is there a problem in the first place. Why higher than 4 rejects is a problem? Against a 15000 pounds/month rate? Is it worth going through the CA process to expect prevention of something that is designed to recur? Or just control the rejects to prevent them from being shipped to customer. When you set "triggers", agree on something that are sensible that when activated will bring the organization added value upon taking action on them.
     
    Qualmx likes this.
  5. Qualmx

    Qualmx Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    443
    Likes Received:
    58
    Trophy Points:
    27
    Location:
    Mexico
    The cases are separated, I defined 4, because of statistics, in the past had 5, so I want to improve it.
    On the other indicator, if suddenly production is very low, of course there is a problem which is addressed by a CA.
    In fact, I think is not rocket science, indicators May be set by taking references of production rates in the past, don't you think like this?
    Thanks
     
  6. tony s

    tony s Well-Known Member

    Joined:
    Sep 10, 2015
    Messages:
    1,222
    Likes Received:
    913
    Trophy Points:
    112
    Location:
    Laguna Philippines
    Agree. If you need to set an improvement objective, you should establish first a baseline. Past performance data should be used for establishing the baseline.
     
    Qualmx likes this.
  7. RoxaneB

    RoxaneB Moderator Staff Member

    Joined:
    Jul 31, 2015
    Messages:
    892
    Likes Received:
    1,033
    Trophy Points:
    92
    Location:
    Ontario, Canada
    Okay, you had 5. Were there common causes? Where they in common areas or in common processes? To lower target with no logical focus doesn't help anyone. As well, if it was 5 in the past year(s), were there actions taken on those 5 (or more) to fix them? By focusing on our unscheduled delays >2 hours, and addressing those issues, we were actively working to lowering the number of occurrences of unscheduled delays. In other words, our actions supported the lowering of the trigger.

    Report both, but I'd go with the % as the metric. The # however provides context to the %.

    For example, in my current position, we have an indicator called referral acceptance. We are contractually obligated to accept a certain amount of work that comes to us from our government funder. We are currently not meeting this target. The thing is the amount of referrals coming our way suddenly increased (with no warning from the government funder), while our staffing numbers remained the same. Our staffing capacity is full. An aging population means more complex care and clients stay on service with us longer, so we are unable to discharge them and free up spots for these new referrals. With a staffing shortage across the province, we are unable to hire more qualified staff.

    So, our % is going down, but our # of accepted referrals has remained steady (even increased slightly at the risk of burning out our staff or failing to provide care to our clients). Since the denominator (i.e., number of incoming referrals has increased) has grown larger, though, our % has dropped despite our workload remaining positive.

    Reporting % is a start to the story, but, fair warning, you may want to be able to give context to it.
     
    Qualmx likes this.

Share This Page