Author Archive

American factories, foreign lands: How can we keep our companies here?

January 25, 2012

Steve Daum

No one will claim that using statistical and monitoring software to stay competitive in the world will reverse the current trend toward manufacturing outsourcing to offshore operations.

But it could help.

A recent New York Times article reported on a challenge that President Obama had made to Apple head Steve Jobs last year, when Jobs hosted a dinner meeting for Silicon Valley luminaries.

Why couldn’t those jobs be brought home? Obama asked.

Lower cost of labor is only part of a more complicated story. Apple, and other manufacturers, need flexible and responsive partners that can, and are willing to, react quickly to changing demands. This more nimble and cooperative supply chain is a valuable asset–a competitive advantage for manufacturers.

“Apple’s executives believe the vast scale of overseas factories as well as the flexibility, diligence and industrial skills of foreign workers have outpaced their American counterparts,” the article asserted. Nearly all electronic goods are now produced in Asia or elsewhere.

If a workforce with appropriate skills is a driving force sending jobs to distant factories then improving American opportunities will require workforce training. Training for workers in statistical thinking, use of statistical tools, emphasis on customer-supplier relationships and shared understanding of quality requirements is an important aspect of this training. Statistical Process Control (SPC) and related quality improvement techniques help to eliminate waste and improve efficiency. These techniques also facilitate communication up and down the supply chain, using common, well-understood concepts.

Have you thought about this issue? What can we do to keep even more companies from taking their manufacturing facilities abroad? Let me hear from you…

Share your charting story, create a charting innovation

October 20, 2010

Steve Daum

You are looking at a chart. You are going through an analysis and interpretation process. What data is being represented? How important is the data? Does the chart signal any changes? Does the chart show anything that is “bad” or “good?”  Does the chart offer proof of quality?

Ultimately, you want to answer the question: is any action required based on what I see?

Now, think about the workflow leading up to this. How did the chart get created?  How was the data gathered? What part of the process was difficult or error prone? Would it have been possible for you to miss this chart among your other tasks?

At PQ we’ve been pondering questions like these for  more than twenty years. We are working hard on our products and services to reduce friction in your quality improvement processes. If you have a charting story to tell, please share it with us; who knows, it may lead to the next great quality improvement solution.

The difference between run charts and control charts

December 3, 2009

Steve Daum

A customer recently asked one of our support representatives the following questions: What is the difference between a run chart and a control chart? And when should I use one vs. the other? These are great questions because they allow us to highlight some of the benefits of control charts.

When you create any chart, you are typically trying to answer a question. For example, you might be asking, “Has my process improved?” or, “Has my process gotten worse?”  You might be asking, “How is the process running today compared to yesterday?” Before you decide on using a run chart or a control chart, consider the type of question you want to answer.

(more…)

The PQ Systems Knowledgebase

October 6, 2009

Steve Daum

Did you know that PQ Systems maintains a robust knowledge base online? It is a repository of common problems and solutions related to using our software and practicing continuous quality improvement. If you are an SQCpack, CHARTrunner, or GAGEpack user, you might just find a solution to your problem here:

http://www.pqsystems.com/kb/activekb/

In a recently-added article, Craig Newland, addresses how to count or group data that has been accumulated during certain periods of time. If you use CHARTrunner, visit: http://www.pqsystems.com/kb/activekb/questions/543/CR%3A+Data+grouping+tardiness to learn how to group your data to make your control chart analysis more telling.

How much did your SPC software cost?

September 10, 2009

Steve Daum

Your first response to this question might be to quote the retail price paid for your SPC software. This would be a factual answer. Fast forward and imagine answering the question after using the software for six months or a year. Would your answer be different?

When evaluating SPC software there are many considerations; price is just one of these. Many other factors will contribute to the full cost of ownership. For example, the software will have to be installed, maintained, and supported. Users will have to be trained. Questions about the software will have to be answered – not just how to use it – but how to apply it in your system.

One of the largest costs might be described as the “make it work in my world” cost. Software designers often have a vision of how work progresses through a system. This vision might or might not align with the way work flows through your organization. I have seen software with impressive feature lists fail to be used – because the workflows just can’t be wedged into the way a firm actually gets work done. A good relationship with an experienced vendor support group can make a big difference – but if the workflows are misaligned – this can prove to be expensive.

Consider also the “manageability” cost. For example, you install the software for several users. They like it, they find it easy use, and they begin using it. However, they each take their own implementation journey. After a few months, important data, and analysis might be strewn about across your network somewhat haphazardly. How will this be managed? Once again, the feature list and price might look impressive – but the long term manageability might increase the real cost of ownership way beyond your initial impression.

It can be difficult to select any software and SPC software is no exception. If you find yourself in this situation, look beyond the initial price and feature list. Take a holistic view. Think about work flows and the long term feasibility of deploying the software and continuing to get knowledgeable support from a reputable vendor. In the end, the retail price might be the least of your worries.

How many charts can you track?

August 21, 2009

Steve Daum

Most of our customers use charts as part of their work. Depending on their job, they may review charts every hour, day, week, or at some other interval. Creating and reviewing these charts is almost never a primary job. Instead, the charts are used as decision support tools; they support the primary job and help a user make better decisions.

Different types of charts have different usage patterns. For example, when looking at a chart of sales, the usage pattern is something like this:

  1. View the chart
  2. Ask this question: Are we on target, above target, or below target?
  3. Depending on the answer, take appropriate action to move towards the desired outcome.
  4. At some later date, repeat.

Statistical process control (SPC) charts have a similar usage pattern with a narrower focus. For example, you might chart the wait time between a customer posting an order and receiving an order acknowledgement.

(more…)

Tips for designing your quality improvement spreadsheets

August 4, 2009

Steve Daum

Among our healthcare customers we find substantial use of Microsoft Excel. A recent survey of CHARTrunner customers found that 68% of them use data in Excel to produce their SPC charts and other analysis related to quality improvement. Excel is powerful and flexible and well suited to this job. However, this power can lead to complicated worksheets that are difficult to use and even more difficult to maintain. Once designed and deployed, a spreadsheet template may be in use for several years. Who will be around to debug an error or correct a formula that is discovered a year later?

To improve the situation requires well designed spreadsheets. Today, spreadsheets are so easy to setup and start to use – that we tend to gloss over using a design process to get started. Here are some tips to think about as you design your next quality improvement spreadsheet:

  1. Decide the primary purpose of the spreadsheet
  2. Make the primary purpose easy to accomplish
  3. Use the simplest possible sheet that accomplishes the purpose
  4. Don’t create future work for yourself
  5. Keep the data “pure”
  6. Be consistent among your sheets
  7. Favor traditional arrangements over weird arrangements
  8. Use a “notes” worksheet to document complex sheets

For a more detailed look at these tips see the following article in the PQ Systems knowledge base:

http://www.pqsystems.com/kb/activekb/questions/541/Excel+tips

Summarizing control chart data with queries

July 6, 2009

Steve Daum

It may seem arcane to some folks – but I enjoy a good query challenge. Recently, a customer needed help with a query to aggregate raw data into a useful form for charting. Since I’ve helped several customer do almost the same thing, I decided to write a short article for our Knowledge Base. Here is the link:

http://www.pqsystems.com/kb/activekb/questions/540/A+simple+query+pattern+for+summarizing+raw+data

If you want to chart data where several rows get summarized for each data point – this article contains example SQL queries. These make good patterns for solving a large class of data summery problems. Don’t let the SQL queries intimidate you – it is pretty simple once you see a few examples. The article includes a download containing example data and example CHARTrunner charts.

Happy querying!

How my thoughts became an article on 4th Generation SPC

June 17, 2009

Steve Daum

I occasionally get the urge to write some thoughts. These often end up as files on my computer; never to be opened or used again. Every few years, I sort by date and start deleting the oldest files. Once in a great while, some document will feel more useful and I’ll develop it further. Eventually, I might show it to someone and say, ‘do you think anyone would read this?’ Usually the answer is no. Here is an exception. The article includes thoughts I’ve developed after several years and many customer conversations about the experience of implementing SPC. Thanks to Barb Cleary and Beth Savage for cleaning this up and finding a publisher, Quality Digest.

http://www.qualitydigest.com/inside/metrology-article/it-s-time-new-and-innovative-approach-spc.html

If you have comments or questions drop me an email at steved@pqsystems.com

Thanks for responding!

June 1, 2009

Steve Daum

We recently surveyed some of our CHARTrunner users. Each time we do this, I am amazed at the diversity among our charting software users. CHARTrunner has found a home in so many different industries. If you responded to the survey, thank you. Your input is invaluable. It helps us to prioritize and plan for future product improvements.

For many years PQ has taken pride that our customers drive the next set of enhancements to our software products. Recently, we began an effort to shorten this feedback cycle. The idea is to get customer input earlier and more often during a project. To this end, we are having many reviews of progress throughout a project. Additionally, we are making daily builds of functioning, if incomplete, new versions as enhancements are added. It is important that we get feedback from non-software developers who have a stake in the project outcome. This means not only sales staff, tech support analysts, and the management team–but most importantly customers.

If you did not have a chance to respond to our survey or if you are interested in providing feedback on early product releases, feel free to drop me an email (steved@pqsystems.com). We would love to hear your suggestions and evaluations of early software builds.

Steve


Follow

Get every new post delivered to your Inbox.