Implementing Statistical Process Control

Lean Learning – Implementing Statistical Process Control

Welcome to the Lean Learning Series on Implementing Statistical Process Control

The Lean Learning Series is a series of sampled notes from my course content on a learning journey to a University of Limerick Masters in Quality Management (Lean Innovation)


Applying ‘process thinking’; it’s possible to differentiate between random causes of variation and special causes in order to better control processes and understand what the target process capability has to be achieved in order to best meet customer requirements.


If satisfied, the desired process capability can be maintained using quality science techniques such as SPC Statistical Process Control to collect data that allows the detection of variation in the process inputs and outputs over time. The data is then utilised to act on ‘Special Causes’ and maintain the processes stability using traditional statistical process control (SPC).


Statistical Process Control (SPC) is a methodology for monitoring a process to identify special causes of variations so that the process can be adjusted when and only when necessary to do so.


Statistical Process Control was developed by Walter A. Shewhart in the early 1920s. W. Edwards Deming later applied SPC methods, initially in the United States during World War II, to the manufacture of military products and other strategically important products. Deming is also credited with introducing SPC methods to Japanese industry in the 1950s.


The SPC chart is fundamentally two trend charts with confidence limits (control limits), one chart for a statistic for centre (mean) and a second chart for a statistic of variation (range or sigma).


From a lean perspective, SPC contributes to reducing waste as well as reducing the likelihood that problems will be passed on to the customer. With its emphasis on early detection and prevention of problems, SPC has a distinct advantage over quality methods such as inspection, that apply resources to detecting and correcting problems in the end product or service.


The process can consist of a number of steps; steps where the product or service change in form are known as process nodes. The sequence of process nodes in a process can be described using a flowchart. Typical flowcharts for manufacturing and service organisations will show hundreds of process nodes. It is therefore critical to identify which are the critical nodes. A critical process node is a step in the process which has a significant impact on the quality of the final product.


Process Capability is a measure of the variation of the process that can be achieved under standard operating conditions of the process. It is a measure of the ability of the process combination of people, machine, methods, material, and measurements to produce a product or service that will consistently meet the specification requirements or customer expectation.


A process is in control when only common cause variation is present.  The goal for any process is to bring it into statistical control by removing any special cause of variation.  Special Cause variation is created by a special event, not normally present in the process, leading to an unexpected change in the process output. The effects are intermittent and unpredictable.


We use process capability studies to give us this understanding as well as a definable measurement of the process.


A seven step approach (there is no definitive approach) to implementing SPC in large and small enterprise.




Step 1. Form implementation teams

Select a ‘Cross-Functional’ Team, vertical and horizontal with

– Process Knowledge

– Production Knowledge

– Product Knowledge

– SPC Knowledge

– Facilitator / Black Belt

Define the Team Roles


Step 2. Identify and rank the critical process nodes

Use Techniques such as Quality Function Deployment to link critical customer requirements and/or product specifications to process specifications

Use a Flowchart to understand the process flow

Start with a node that has a high probability of success

Develop an implementation plan


Step 3. Develop process flow for the critical node

Develop a detailed process flow showing:

Process Operations

Data Collection Points

Storage Points

Wait Points

The chart is to represent what actually happens, not what the team thinks happens or would like to happen! So go to the genba!


Step 4. Prioritise Critical Outputs and Inputs


Consider the following when deciding your critical outputs:

Customer Requirements / Product Specifications


Product and Process Knowledge

Design of Experiments

Product and Process FMEA’s

Choose Variables rather than Attributes


Conduct a Cause and Effect Analysis to establish the effect of inputs on the critical outputs selected in Step 4

Design of Experiments can also be conducted to determine the effect each input factor has on the centre of location and causes of variation in the output



Step 5. Establish Process Capability

Lock critical inputs

Lock the critical inputs if possible to minimise causes of variation in the output; i.e. set process control limits on the inputs outside which the process will never run

Determine measurement capability

Conduct a measurement capability study; there is no point measuring process capability unless we know what our measurement capability is

Consider both repeatability and reproducibility.

If measurement capability is more than 10% of the expected process variation then improvement is needed in your measurement capability before you proceed

Conduct a Capability Study

Initially focus on one run of product with as many input factors locked as possible

If the Cp > 1 then the process is run under normal operating conditions taking samples from each lot until you have 30 – 50 points



Step 6. Implement Control Charts and OCAPs

Calculate all the limits for the Control Charts. The procedure for building Control Charts for both Variables and Attributes will be discussed in the next units.

Before Control Charts are introduced on-line everyone involved should be trained in how to record the data, complete the control chart and read the control chart

Provision should also be made for what to do when the process is ‘Out-Of-Control’.


When a process can run or when it should be stopped

What action should be taken by the Control Chart Operator in an ‘out-of-control’ situation?

What and Where to document the situation



Step 7. Continuously Improve Cp’s

SPC can be used in detect and correct mode to maintain process quality levels

By using the control charts to analyse and improve the process, SPC can be used to continuously improve the process


During ongoing production, If we detect that the process is out of statistical control, then we do not necessarily stop the process. We investigate and determine the source of special cause variation. Once we find the root cause we then eliminate it so that it should not recur.  SPC should be viewed as a management tool and system, not just as a set of control charts.


If you would like to read more in the Lean Learning Series please read

Lean Learning Six Sigma



Comments are Closed