Continuous Improvements with Control Charts

Bizmanualz Solutions

In the previous series of articles, we discussed applying Statistical Process Control (SPC) to business processes. Although SPC has been around for a long time and is applicable to the current world of Lean, Six Sigma, and ISO, it is often overlooked.

SPC is a great tool to not only monitor the current and recent behavior of a process, but is also a great tool to support Continuous Improvement activities. Our previous article was about control chart, which is, in simple terms, graphically plotted SPC information. The most typical among control charts is the process average and range Control Chart, commonly called the X-bar and R chart. This type of data is measured or variable data, as opposed to attribute type of data. Please note that there are other types of Control Charts for attribute data.

“Control without action is simply a hobby!�?
-Kauru Ishakawa

Average and Range Control Charts

To plot the process average (X-bar) and range (R) Control Chart, samples of product are obtained, usually at specified intervals, and measured. The sample size should be predetermined and then maintained. Typical sample sizes are subgroups of 3, 4, 5, or 6 pieces, with 5 being the optimal. As the process is running, the samples, of either in-process or finished product, depending on what you are interested in controlling, are randomly obtained and then measured for the required characteristics. These measures are entered into the company SPC data collection system. This activity is repeated at the specified sample interval, say hourly, for as long as the process is running.

There are many types of software programs available for recording data and creating Control Charts, or for those eccentrics out there, one can calculate their own data points, process averages, ranges, and so forth. However, most of us will use software, which will calculate our subgroup average, subgroup range, process average, process range, control limits, standard deviation and process capability. “Wow! The software can provide a lot of information, but how is it used,�? you may ask?

Control Charting Software

As the subgroup measures are entered into the system, a whole lot of calculations are performed and plotted or listed on the Control Chart. The chart and data is not very useful until a sufficient number of data points (subgroups) have been entered into the system and plotted. 25 subgroups will start to give one a good “picture�? of their process, but as time passes and hundreds, and then thousands of subgroup data points have been entered and calculated, the “picture�? of the process becomes even more accurate and clearer. One of the nice things about SPC software programs is that one can isolate any period of time, such as a specific shift, and display that time period’s specific Control Chart and other data.

Once a sufficient amount of data has been entered and a good picture is obtained, several pieces of very useful information are available. First of all there is a graphic “picture�?. Even an SPC novice or untrained operator will be able to tell from the “picture�? if there has been an event with an assignable cause or possible trends. The more experienced and trained operator, manager, or engineer will be able to exam the process spread, as identified by the Control Limits and the standard deviation or sigma, to determine if the process is “in control.�? Then Control Chart graph can be observed for any subgroup trends to see if there is a high possibility of going “out of control.�?

Continuous Improvements

The Continual Improvement team will look at these Control Limits and process variation to see if there may be opportunities for reducing this routine or common cause variation. As for myself I like to first glance at the picture, and then focus on the Process Capability Index or Cpk, which is a numerical index displayed to the side of the Control Chart Cpk for the process is calculated each time data from a process is entered into the system.

If I were to call my friend Mike and ask him how his production line #3 is doing today and his answer were; “Steve we’re currently running a Cpk of .63.�? My response would be; “Mike, you’ve obviously got some work to do…I’ll give you a call next week!�? However, if Mike’s answer was; “Steve, line #3 is running at a Cpk of 1.4,�? my response would be; “Great! Let’s meet for lunch today and you’re buying!�?

One thing to know is that there is a lot of misunderstanding of Cpk. Simply put, it is a measure of the processes capability of meeting the product specification that is being measured. Another way to look at it is, are the processes calculated control limits inside or with in the specification limits?

“It is a capital mistake to theorize before you have all the evidence. It biases the judgment.”
-Sir Arthur Conan Doyle

Regardless of your level of SPC understanding or your understanding of process variation, SPC and Control Charts will provide you with valuable information on your processes. If you are not sure if you’d be interested in better understanding your process variation and using SPC consider these statistical charts on the right.

Statistical Process Control Training

If you do not have a thorough understanding of what is depicted in each figure, and you want to understand and control your process variation, then you should consider taking a Statistical Process Control (SPC) class. As fate would have it, Bizmanualz is offering just that – a two-day Statistical Process Control Training class.

In this class you will receive a quick basic review of Quality Control, followed by the development of a mini factory. In this factory you will produce product in real-life scenarios, complete with problems and management interference. Through the measurement of your product and the accumulation of data, the concept of product and process variation will be explored and understood. Then the data from the mini factory product will be used to understand the various SPC terms and a Control Chart will be developed. The capability of the mini factory process to meet specification, Cpk, will be calculated.

Finally, to fully understand SPC and variation, problems are worked through in class to calculate the Control Limits, standard deviation, Cp, and Cpk. Upon completion, the student will have a basic understanding of process variation and process control. Additionally, the student will understand when to and not to take action on a process, as well as to identify opportunities for improvement. This class is suited for the process operator, engineers, and managers. As a prerequisite, prospective students should understand how to convert fractions to decimals, how to use a ruler to measure at least to eighths, and how to use basic calculator functions.

“the significant problems we face cannot be solved at the same level of thinking we were at when we created them.�?
-Albert Einstein

Learn more about developing policies, procedures and processes, or about improving your organization by attending the next How to Create Well Defined Procedures and Processes or Statistical Process Control classes. TO address other training needs, please visit the Bizmanualz Training Website.

Sphere: Related Content