iSixSigma

AlexanderSO

Forum Replies Created

Forum Replies Created

Viewing 11 posts - 1 through 11 (of 11 total)
  • Author
    Posts
  • #168045

    AlexanderSO
    Participant

    Hi, sorry if I´m late to join the talk.  I suggest you use a MIL STD to definie your sample policy, the first step will be to define the type and level of inspection you want to perform and that depends on the requirements of your process, for example you choose a severe inspection when your process had shown bad performance according with the quality reqs, and a reduce inspection when the performance had been good (in the MIL STD 105D you can choose between severe, normal and reduced type), also you choose a level (I,II,III) and the selection allows to change the size of the sample without affecting the producer risk but changing the risk of the customer and the decision must be made knowing the effects and cost of each risk, in addition to this you must define an acceptance level for the production lot, the initial recomendation given in the literature is to use an acceptance of 95%, but it depends on the nature of your process.
    Over the selected sample you can perform quality and capacity analysis and estimate your sigma level and define action plans to achieve the desired level.

    0
    #167056

    AlexanderSO
    Participant

    Hi, usually star with a practical exercise (simulation type) applying Lean methodology, this exercise is very helpfull to achieve a relaxed enviroment and motivate the audience, and then star building the Value Stream Map giving the people the opportunity to explain the activities performed by them to rest of the group, and once the VSM is completed with the time line and value zones identified, as a group point the opportunity areas.

    0
    #166352

    AlexanderSO
    Participant

    Hi, I believe you are taking my comment a little bit to far, I´m not saying that the inspection procedure is unnecesary, it is requiered because is the stage when a lot of the defects are being detected, the way this inspection is performed depends of a lot of things related with the process, check list or pokajoke like in Toyota or many other ways used in industry.  What I´m trying to say is that a quality problem should not be adressed adding muscle to the inspection line, this effort wiil be more usefull aplying it on the source of variations or root causes, what Toyota did was creating a strong problem solving way of work and when the line was stopped they get around and look for the root cause and address it, this approach is not agains the use of smart statistics to monitor and address actions over a process, the feeling about the statistical approach is than in most industrial enviroments the solutions take too long to be implemented and much of this time have been asigned to the requiered time to perform statistical analysis (sometimes unneeded ).  But if you take a look at the industrial history of Japan since the end of the second world war, most of the success in bussiness dynamics and product quiality achieved by Japan industries is due to the intervention of Dr. Edwards Deming and his satistical approach to problem solving, and until today the Deming medal award is one of the most prestigious awards that a Japanese industry can obtain.

    0
    #166088

    AlexanderSO
    Participant

    Bill, I think that the problem should be addressed with an statistical approach, and insteed of putting extra effort on inspection you should find and minimize the sources of variation at the source of the defects, in the long term it makes no difference to improve the inspection method allowing you to catch all the defective units, at the end the cost of bad quality still be high, you will no have claims from customers but you´ll have a lot of rejected units waiting for overprocess or to be scraped, and this will not give real benefit to the bussiness.

    0
    #165346

    AlexanderSO
    Participant

    I think you can use the MIL STD to define the sampling of your process, in the MIL STD you can choose from a wide range of sampling options based on the acceptance that you requiered to guarantee you quality level.

    0
    #164846

    AlexanderSO
    Participant

    Hi Darrel, I suggets that before starting an argument with your boss check the end user, for this you can use historical data that tells you if the parts over the 0.004 spec value performed well or not and if they did maybe tou need to validate the upper limit of the spec.  Also if the extra digit in the measure really helps you to control the quality of the product or is irrelevant to the end use performance.
    On the other hand if they didn´t perform as expected you can explain with data on hand your point.

    0
    #164845

    AlexanderSO
    Participant

    Hi Rob.
    I understand from your post that your final customer isn´t getting defective parts, however following the product road just before the final customer is the quality audit, and this guys are receiving defective products, so you can use them as your project customer.
    With them you can build a QFD matrix (Quality Function Deployment or usually known as the quality house) to try to obtain a relation between the reject causes they use and the variables of the up stream processes.

    0
    #164085

    AlexanderSO
    Participant

    Hi, if your main focus is to determine if the various set of data fits one to each other you can use a two step method.
    The first step is to check the difference between the series usig min square error, for each data calculate as follows (reference value-evaluated value)^2 then at the end add all the values and divide them by the number of data evaluated and this number represents the difference between the series.
    The second step is visual, you can plot the series one over each other and check if the behavior is the same.

    0
    #163838

    AlexanderSO
    Participant

    Hi.
    When you improve the on time shipments you are impacting several fronts.
    The benefits inside your company could be less space requiered for inventory (due to an increased exit rate, you can quantify this using the typical formula used to define EOQ models, also the free space now could be used for additional money generating activities), increase in the revenue (due to a reduced number of orders partialy shipped or not shipped at all that the customers usually cancel or close them, for which you can´t collect the total initial amount of money only the amount representing the real shippment), if you align your productive system to on time deliveries you will star to have productivity improvements opportunities that may lead to additional sales.
    Viewing from the outside on time shippments puts your company in a stronger position in the market, due to your improved reliability the customers will put your company at the first buying options (you need to validate this with your sales force to see if the market you are attending is sensitive about delivering time) and this on the long run will give your company a major market share and will create loyalty among your customers. 

    0
    #156048

    AlexanderSO
    Participant

    Hi, I just want to add one thing, I have seen your explanation of how you run your replenishment system, and is great, because you are using the effect of the variation, and it allows you to keep syncronized with the demand changes.
    This follow up of the demand is one of the reasons to recalculate the Takt Time every period, the other main function of knowing the actual Takt Time is to constantly compare it with you Cycle Time and be sure that your facility or process is capable of following the demand and waht adjustments you need to do to do it, for example when your cycle time is bigger than the requiered Takt Time, you can decide to use outsourcing, extra time or create temporal inventory, in the other hand when Takt Time is bigger you can decide to schedule personnel vacations, miantenance or other activities that slow your process a little.

    0
    #155715

    AlexanderSO
    Participant

    Hi, I have some thing to add to the already said.
    Is very important to use the effect of the variation in the demand, however you can’t loose sight of the maintain cost.  For this reason I thing one should direct the efforts in a tactical manner, the method I use is this.
    1. Do an ABC clasification of your products (I use Paretto’s percentages A for the 80% B for the next 15% amd C for the last 5%), I do this with the last 12 months and review it every 4 months.
    2. Define the service level for each clasification, what I call the service level is the amount (percentage) of the normal distribution you want to cover, the 50% of the coverage is achieved with the aritmetic average, and is not really much advantaje in going to near the upper limit because the beneffit of a major inventory level is to low because you need much increase in your stock and gaing very few points in coverage, the values I use are 90% for A, 70% for B and 40% for C.
    3. Calculate the corrected average, what I call the corrected average is the value requiered to obtain the coverage of the normal distribution, as an addition you can check if your data fit into a normal distribution (you can use the asymetric factor and the cortuosis).  This is the inventory level that you shuol attain.
    4. Determine the lot size and the frecuency of the orders, I thing using the EOQ (Economic order quantity) as an initial approach is a good star, this is a very old theory, but it uses the cost factor, and this factor will always be important in any bussiness.
     
    Please let me know if this is usefull to you.
    Best regards.

    0
Viewing 11 posts - 1 through 11 (of 11 total)