12

This gives you the starting point for a return on investment calculation. It tells you the value in fixing each issue in terms of how many defects would be prevented. The idea is that the curve shows the cumulative percentage of issues vs. the total issues. I've drawn a red cross-hair at where the curve crosses the 56.8 mark - 56.8 is 80% of the 71 issues. ...


9

TL;DR: It depends, but probably not You might argue that "all 15 samples are within spec, so the supplier should be approved". Not so fast: depending on the parameters of your full production run, the 15 samples may or may not be statistically significant. There are numerous calculators online to do these calculations; I used this one. A good resource for ...


6

There are two different aspects in your measurement. On the one hand, you are dealing with tolerances. On the other hand, you cover probabilities in measurement systems. Just for a rough calculation: The probability of the real length to be within 9.8mm and 10mm is 95%. The certainty of this measurement depends on the distribution of your probability. For ...


5

First off, always remember that garbage in = garbage out; so if your data is garbage then your statistics will be garbage. In this situation your optimal data would be something like Run Hours Until Failure and your entire dataset would have failed already. With this in mind you may want to choose a conservative number from whatever statistic you calculate....


4

Measurement variations are very common and should be taken in to consideration when engineering systems. In most cases high precision equipment is available but might be cost prohibitive to justify purchasing for the project. Therefore, the goal of the engineer is to design the system to account for measurement variation. In this case the min and max limits ...


4

If you don't have hard data, making assumptions (preferably "reasonable" ones) is the only option you have. (Maybe that's why engineers used to call their slide rules "guessing sticks...") You can't ignore the fact that most of the units have not failed so far. A plausible approach to this would be to use the times-to-failure that you know, to fit the ...


4

Based on the discussion in comments, you probably want to use a bar chart combined with a line for the error. Have a look at Create a combination chart for more details. In your case, and using dummy data, it would probably look something like this: Data in Excel Resulting combination graph


4

If you know that the errors in your measurements are truly random and have a normal distribution then you can improve the resolution of a measurement by taking several measurements and averaging them. This technique is called oversampling and is common in digital data acquisition systems where the hardware resolution may be low. You mentioned accuracy, ...


3

The only way to determine the accuracy to which any measuring device provides measurements is to calibrate it against a device of known accuracy and known errors for measurements. You technique is partially correct; don't just do the error measurement for the limits of the device as one population or sample bin. This is because measurement errors are not ...


3

Per @Dan's suggestion, I googled "nomography" and found an article titled "The Lost Art of Nomography" by Ron Doerfler, published in the UMAP Journal 30.4 (2009). The article describes a compound-nomograph, which is pretty much what I need. Here is a screen shot of such a device, taken from the article: The image shows a nomograph for three independent ...


3

You do not say what sort of capacitors are involved. Tests will vary with capacitor type. Electrolytics can be tested for eg leakage, ESR, behaviour as they approach rated voltage, ... . Ceramic capacitors that are "of unknown origin" may well exhibit temperature dependant variations not matched to they specifications. [Tantalum capacitors may explode with ...


3

The average and standard deviation from the above sample set are: Average ($\mu$): 4.754V Standard Deviation ($\sigma$): 0.142368 Mean: 4.74 USL: 5.5 LSL: 4.5 C$_{p}$ : 1.1706 C$_{pl}$: 0.5947 C$_{pu}$: 1.7466 C$_{pk}$: 0.5947 Equations $\text{Pr}(\mu - \sigma \leq x \leq \mu + \sigma) \approx 0.6827$ $\text{Pr}(\mu - 2\sigma \leq x \leq \mu + 2\sigma) ...


3

Quite seriously: Friend Don't Let Friends Use Excel. Learn to use Matlab/Octave, or python, or R, or just plain Gnuplot. No matter what tool you use, since "significance" is a separate variable, you need to convert your collection of "*" , "**" etc to numeric values and plot the desired symbols at the matching category locations.


3

Just a quick summary of my concluding thoughts on this step in your research. I understand that you have been advised to use the Sigmoidal Function, so this discussion will try to explain it's potential application. First, I don't think that an Artificial Neural Network (ANN) would apply to your problem. Typically ANN would consider a range of factors on ...


3

If we talk about quality and 6-sigma (which roughly means 1 ppm defects), the 1 ppm refers to test escapes, not test rejects. Even if a part only costs \$1, if testing fails to catch a bad part and we accidentally send it to a customer, where it fails in the field, the eventual cost (in failure analysis, corrective actions, documentation, time spent ...


2

I agree with What Trevor said in a comment. You specified 5V ±10% (4.5-5.5 V) and that's what you got. I don't see the problem here. All the samples you got are within spec. There may be other reasons to not use this vendor, but samples failing to meet specs is not one of them. However, first article samples don't tell you that the vendor is OK, ...


2

This actually isn't as much of an engineering question as it is a physiology question. There are actually a number of widely used estimates to predict your one-rep maximum, aka "1RM", if you know how many repetitions you can do at a lower weight. See here for more info. All of the methods are based on empirical studies, and are basically look-up tables of ...


2

I don't know of any mathematical or statistical techniques that would help you analyse machine utlization data. From the additional information you've given in the comments, my understanding is that your aim is the minimize the cost of keeping spare parts for the machines based on machine usage. The critical factor is going to be the duration of usage for ...


2

Looking for good and reliable suppliers might be one of best solutions. Vendors such as Digi-key, mouser, Newark always supply high quality electronic components. But understandably they come at a higher cost. If directly purchased from a manufacture it might be good idea to request for test data from suppliers to validate that product is meets specification....


2

In the mechanical world, if you have a million bars, and pull on them until they yield, you'll get a wide variety of failures. These failures, if you counted them with how many failed at what strength, you usually end up with a normal distribution, though to make the math easy we usually fit it into a log-normal distribution. When dealing with a part who ...


2

Your approach is broadly correct. If you are only interested in the accuracy of your system you probably want to use something like the maximum error. Your accuracy is then +/- Max error with the assumption that real errors are uniformly distributed within this range (a uniform distribution will often be an overestimation but is a simple option when no ...


2

The statistical tool Weibull as suggested by the previous two responded is the tool of choice for Mean Time To Failure (MTTF) calculations. Based on your comment as capture below, it appears that Weibull Analysis didn’t generate expected results. Most statisticians that I have worked with recommend a sample size of 30 for most statistical analysis. My ...


2

In the first order the most commonly used quality engineering tools are average, standard deviation, histograms, pareto charts, control charts. In the second order tool such as Design for Experiments (DOE), ANOVA, student t-test, z-test, confidence testing, gauge R&R, Ishikawa diagram most likely are next set of tools that are most used. From software ...


2

Get an ASQ certificate: ASQ certification is a formal recognition by ASQ that an individual has demonstrated a proficiency within, and comprehension of, a specific body of knowledge. They have a bunch of different certificates that are applicable to the different engineers/techs you guys have: Certified Quality Engineer (CQE) Certified Quality ...


2

The point where center of mass lies so the center of gravity. So, for center of mass For rod AC. B is the mid point of AC.So,let us take A as origin So coordinates of B($\frac{l}{2}$,0).Similarly for midpoint of BG We say there are four divisions which are equally space hence the length of BD,DE,EF,FG are$\frac{l}{4}$ So the mid point is E the mid point of ...


2

It appears that your dataset has integer row and column header values starting at 0. You can try the following code. Turn on Excel's Developer ribbon if it's not already on. Click the Visual Basic button. Insert | Module. Paste the code below into the module. Adjust the rowOffset value to suit. Enter the formula =Interpol2D(y, x) into your spreadsheet where ...


2

Search for daily maximum and minimum temperatures for your local region. If there is a river nearby, search for river level or river gauge height. If it rains, search for annual rainfall in your region.


2

Data is data. This is the capability of the design and manufacturing process. Trust the data. If there is a need to improve the quality of the data one could take for example 3 measurements for each sample and then take the average. So if there is a concern of quality of the data collection, this method will help minimize the error, with regard to the 22.1kN ...


1

Out of desperation, I got in contact with some very nice statisticians at NIST. They walked me through the answer to my question, which boils down to: control charts are an empirically derived chart with strong statistical backing. However, because of this by its very nature, there is no one right answer. It depends on how historically stable your process is,...


Only top voted, non community-wiki answers of a minimum length are eligible