Using Statistical Process Control To Identify Variation

Using Statistical Process Control to Identify Variation

SPC statistical process controlA robust system of metrics is significant not just to analyze progress during a lean Six Sigma transformation, but can aid in identifying the source of variation and waste. Isolating the cause of such variations greatly simplifies effectively re-defining processes, enabling organizations to perform and produce in an improved manner.

While a host of tools and measures are typically used within a lean Six Sigma program, one in particular, Statistical Process Control or SPC, examines processes as well as identifying sources of variation. SPC applies statistical methods to achieve this task.

Within Statistical Process Control, data is statistically compared within the context of its occurrence. For example, factors such as shifts, operators and production events would be included in an SPC study within a lean manufacturing Six Sigma program. Data is plotted within the confines of specific zones and rules to represent these contextual factors accurately. As a result, the information is easily read, greatly simplifying the task of monitoring and acting upon the information.

Statistical Process Control Methodology

SPC follows a set process known as DMAIC. This acronym represents the stages Define, Measure, Analyze, Improve and Control. Common activities during these stages can include:

  • Define – benchmarking, process flow mapping, flowcharts
  • Measure – defect metrics, data collection, sampling
  • Analyze – Fishbone diagrams, failure analysis, root cause analysis
  • Improve – modeling, tolerance control, defect control, design changes
  • Control – SPC control charts, performance management

Statistical Process Control Charts

From the efforts detailed above, the generated data is then plotted on a control chart. This chart illustrates the statistical stability of a particular process. Variations within the process over time stand out clearly within this context. Control charts also make it possible to distinguish between normal variations and those warranting investigation.

There are a number of control charts that may be used for SPC. Frequent choices include:

  • Fishbone Diagrams
    These are also known as herringbone diagrams, cause-and-effect diagrams or Ishikawa diagrams, Fishbone diagrams are best for problem-solving, as they can illustrate the relationships between different variables.
  • Frequency Histograms
    This is a basic SPC tool that conveys the average or mean of the data, the variation present, variation patterns, and whether or not the process is within specifications.
  • SPC Run Charts
    Run charts can be particular revealing. They describe process characteristics over time. As a result, they can reveal relationships between variables.
  • Pareto Charts
    Pareto charts identify the most frequent process factors. They can be an aid in identifying the best use of limited resources by identifying the most significant variables warranting intervention.