Featured Product
This Week in Quality Digest Live
Statistics Features
Donald J. Wheeler
What they forgot to tell you in your statistics class
Donald J. Wheeler
How does it compare with a process behavior chart?
Donald J. Wheeler
What difference does the probability model make?
Donald J. Wheeler
What is the link between the indexes and the models?
Donald J. Wheeler
What do predictable processes have in common with chaos?

More Features

Statistics News
New capability delivers deeper productivity insights to help manufacturers meet labor challenges
Day and a half workshop to learn, retain, and transfer GD&T knowledge across an organization
Elsmar Cove is a leading forum for quality and standards compliance
InfinityQS’ quality solutions have helped cold food and beverage manufacturers around the world optimize quality and safety
User friendly graphical user interface makes the R-based statistical engine easily accessible to anyone
Collect measurements, visual defect information, simple Go/No-Go situations from any online device
Good quality is adding an average of 11 percent to organizations’ revenue growth
Ability to subscribe with single-user minimum, floating license, and no long-term commitment
A guide for practitioners and managers

More News

Eston Martz


Which Supplier Should You Choose?

Check the data

Published: Tuesday, September 29, 2015 - 13:12

Whatever industry you’re in, you’re going to need to buy supplies. If you’re a printer, you’ll need to purchase inks, various types of printing equipment, and paper. If you’re in manufacturing, you’ll need to obtain parts that you don’t make yourself. But how do you know you’re making the right choice when you have multiple suppliers vying to fulfill your orders? How can you be sure you’re selecting the vendor with the highest quality, or eliminating the supplier whose products aren’t meeting your expectations?

Let’s take a look at an example from automotive manufacturing to see how we can use data to make an informed decision about the options.

Camshaft problems

Thanks to camshafts that don’t meet specifications, too many of your company’s engines are failing. It’s harming your reputation and revenue. Your company has two different camshaft suppliers, and it’s up to you to figure out if camshafts from one or both of them are failing to meet standards.

The camshafts in your engines must be 600 mm long, plus or minus 2 mm. To acquire a basic understanding of how your suppliers are doing, you measure 100 camshafts from each supplier, sampling five shafts each from 20 different batches, and record the data in a Minitab worksheet.

Once you have your data in hand, the Assistant in Minitab Statistical Software can tell you what the data say. If you’re not already using our software and you want to play along, you can get a free 30-day trial version.

Step 1: Graph the data

Seeing your data in graphical form is always a good place to start your analysis. So in Minitab, select Assistant > Graphical Analysis.

The Assistant offers you a choice of several different graph options for each of three possible objectives. Because we’re not graphing variables over time, or looking for relationships between variables, let’s consider the options available under “Graph the distribution of data,” which include histograms, boxplots, and Pareto charts, among others.

Click here for larger image.

A basic summary of the data is a good place to start. Click the Graphical Summary button and complete the dialog box as shown.

The Assistant outputs a diagnostic report, a report card, and the summary report shown below.

Click here for larger image.

The data table in this summary report reveals that the means of the camshafts sampled from supplier 2 and supplier 1 are both very close to the target of 600 mm.

But the report also reveals a critical difference between the suppliers: Although neither data set contains any outliers, there is clearly more variation in the lengths of camshafts from supplier 2.

The supplier 1 distribution graph on the left is clustered tightly around the mean, while the one for supplier 2 reflects a wider range of values. The graph of the data in worksheet order shows that supplier 1’s values hew tightly to the center line, compared to supplier 2’s more extreme up-and-down pattern of variation.

Returning to the table of summary statistics at the bottom of the output, a check of the basic statistics quantifies the difference in variation between the samples: the standard deviation of supplier 1’s samples is 0.31, compared to 1.87 for supplier 2.

We already have solid evidence to support choosing supplier 1 over supplier 2, but the Assistant makes it easy to get even more insight about the performance of each supplier so you can make the most informed decision.

Step 2: Perform a capability analysis

You want to assess the ability of each supplier to deliver camshafts relative to your specifications, so you select the Capability Analysis option in the Assistant menu.

The analysis you need depends on what type of data you have. The measurements you’ve collected are continuous. The Assistant directs you to the appropriate analysis.

Click here for larger image.

Clicking the “More” button displays the requirements and assumptions that need to be satisfied for the analysis to be valid.

You already know your data are reasonably normal, that the data have been collected in rational subgroups, and that you have enough for reliable estimates—we saw this in the initial Graphical Analysis.

However, the Assistant also notes that you should collect data from a process that is stable. You haven’t evaluated that yet, but fortunately, the Assistant automatically assesses process stability as part of its capability analysis.

Confident that your data were collected appropriately, you click the Capability Analysis button and complete the dialog box as shown to analyze supplier 1:

Click here for larger image.

The Assistant produces all the output you need in a clear, easy-to-follow format. The diagnostic report offers detailed information about the analysis, while the report card flags potential problems. In this case, the report card verified that the data were from a stable process, and that the analysis should be reasonably precise and accurate.

Click here for larger image.

The summary report gives you the bottom-line results of the analysis.

Click here for larger image.

As shown in the scale at the top left of the report, supplier 1’s process is very capable of delivering camshafts that meet your requirements. The histogram shows that all of the data fall within specifications, while the Ppk measurement of overall capability is 1.94, exceeding the typical industry benchmark of 1.33.

Now you perform the analysis for supplier 2. Once again, the report card verifies the stability of the process and confirms that the capability analysis should be reasonably accurate and precise. However, the Assistant’s summary report for supplier 2 reveals a very different situation in most other respects:

Click here for larger image.

The scale at the top left of the report shows that supplier 2’s ability to provide parts that meet your specifications is quite low, while the histogram shows that an alarming number of the camshafts in your sample fall outside your spec limits. Supplier 2’s Ppk is 0.31, far below the 1.33 industry benchmark. And with a defect rate of 28.95 percent, you can expect more than one-quarter of the motors you assemble using supplier 2’s camshafts to require rework.

You can use the Assistant output you created to explain exactly why you should continue to acquire camshafts from supplier 1. In addition, even though you’ll discontinue using supplier 2 for now, you can give this supplier important information to help it improve its processes for the future.

A clear answer

Whatever your business, you count on your suppliers to provide deliverables that meet your requirements. You have seen how the Assistant can help you visualize the quality of the goods you receive, and perform a detailed analysis of your suppliers’ capability to deliver quality products.


About The Author

Eston Martz’s picture

Eston Martz

For Eston Martz, analyzing data is an extremely powerful tool that helps us understand the world—which is why statistics is central to quality improvement methods such as lean and Six Sigma. While working as a writer, Martz began to appreciate the beauty in a robust, thorough analysis and wanted to learn more. To the astonishment of his friends, he started a master’s degree in applied statistics. Since joining Minitab, Martz has learned that a lot of people feel the same way about statistics as he used to. That’s why he writes for Minitab’s blog: “I’ve overcome the fear of statistics and acquired a real passion for it,” says Martz. “And if I can learn to understand and apply statistics, so can you.”