Statistical Process Control (SPC) is an important statistical quality control tool. Its concept originated from the manufacturing industry, but is now applicable for analysing, controlling, and improving any kind of repeatable process.
In general, the project processes might not be amenable to SPC techniques because, by definition, the projects are of temporary nature with a fixed start and end dates, and result in a single project outcome. Therefore, the processes of a project are slightly different from the operations in a manufacturing industry, where identical products are mass produced.
Irrespective of this, you can deploy SPC techniques in projects organizations that use similar kinds of processes in every project. The project managers must know about the use of SPC techniques for sourcing components and for repeatable processes.
SPC theory
In any process where identical products are produced repeatedly using the same set of operations over a period, the concept of SPC is applied more easily. Whenever a product is manufactured in high volumes, the objective of the production process is to ensure that all the products meet the stated customer specifications.
If a customer specifies that the product should have a dimension of x measurement units, then every product manufactured should be of x measurement units. However, this is not feasible in real life.
In this scenario, you can design a process in such a way that the final product is of x measurement units. The quality of the process is its ability to create products of exact dimensions repeatedly and consistently. The output of this process depends on the following factors called the five M’s:
Methods: The efficiency and consistency of process steps involved
Machines: The accuracy of equipment and their performance
Manpower: The ability of workers to produce the product
Measurement: The accuracy of the measurement process
Materials used: The quality of the input resources
If a project manager controls the process to create products of exact dimensions over a period consistently, then he/she can ensure quality of all the above five factors.
According to the SPC theory, in the above process where all the M’s are completely controlled, there will still be variations in the product dimensions but these variations will follow a defined pattern. This is the pattern of a normal distribution.
A normal distribution is a symmetrical distribution of a product dimension around the mean or average dimension in the shape of a bell curve.
