The purpose of quality control is to ascertain whether the quality of a product or service is being maintained or if there is any variation in size, weight, finish etc. In every production process, there are always some standard specifications laid down either by the producer or the consumer.
A good quality item is one which conforms to these specifications. However, variation in the quality of a product is inherent in every production process due to a number of factors. So, it is necessary to ascertain the variation which may be quantitative and qualitative.
Quantitative characteristics are those which can be directly measured, e.g. weight, height, diameter etc. and such variations can be noted with the help of specific instruments. On the other hand in qualitative characteristics, direct quantitative measurement is not possible, e.g. cracks, breakage, colour etc.
These can be determined by inspection only or by distinguishing between defective and nondefective items. But variation in the quality of products being an inherent characteristic of manufacturing process, irrespective of all possible precautions and measures, there are possibilities of random disturbances responsible for deviations in the quality of the product from the set standards.
The sources of these disturbances are known as chance causes, e.g. changes in machine speed due to sudden changes in temperature or voltage of power supply etc. The presence of these causes in the system is due to multitude of reasons which are difficult to identify and uneconomic to eliminate. There may be other sources of variations which further cause the product to deviate from set standards. These causes are individual and can be identified and eliminated economically. The magnitude of variability due to these causes varies with the conditions of the production process, nature of raw material, behaviour of operation etc. These causes are known as assignable causes.
Statistical quality control refers to the technique of ascertaining whether the variation in the quality of the product is due to chance causes or due to assignable causes. If the variation is due to assignable causes, it is detected and some corrective action is planned to improve the quality of the product.
Statistical quality control is carried out with the help of control charts. To prepare a control chart the whole production line is divided into a number of sub-groups. The basis of selecting these sub-groups is such that variation in the quality of items within each sub-group is attributed due to chance causes, whereas the corresponding variation between various sub-groups can be due to assignable causes.
The variation of a quality characteristic within and between the sub-groups is analysed by some method to identify whether the process is in control or not. Briefly speaking, statistical quality control is based on statistical estimation of errors or possible variation from the average (normal) proportion of errors. In its simple operation, it involves specifying the quality levels and limits on control and then plotting the variations.