Statistical Process Control (SPC) is a scientific, data-driven methodologymethodology for measuring and controlling quality during the manufacturingprocess. Quality data in the form of Product or Process measurements areobtained in real-time during manufacturing. This data is then plotted on agraph with pre-determined control limits. Control limits are determined by thecapability of the process, whereas specification limits are determined by theclient’s needs. Statistical Process Control (SPC) is also said to be a powerfulcollection of problem solving tools useful in achieving process stability andimproving capability through the reduction of variability, Montgomery andRunger (2007) .
It is a technique used to determine whether a process is in statisticalcontrol.In most manufacturing and non –manufacturing systems one of the toolsthat has been widely used for Quality Control is the Shewhart’s StatisticalProcess Control Charts. Shewhart charts are typically used to distinguishbetween variations due to special causes from variations due tocommon causes. Special causes are referred to as assignablecauses like sporadic problems suchas the failure of a particular machine or a mistakenlyrecorded measurement and these are identifiable and correctable .Common causes are problem inherent in every process andsomehow cannot be avoided. The process issaid to be in statistical control when the special causeshave been identified and eliminated, andonce statistical control has been established, Shewhartcharts can be used to monitor the processfor the occurrence of future special causes and to measureand reduce the effects of commoncauses, Montgomery (2005). These charts have been used andare still being used to monitor process performance ,Process monitoring alsoplays a key role in ensuring that the plant performance satisfies the operatingobjectives.
The general objectives of process monitoring are: Routine Monitoring-Ensure that process variables are within specified limits. Detection andDiagnosis. -Detect abnormal process operation and diagnose the root cause.Preventive Monitoring.-Detect abnormal situations early enough so that correctiveaction can be taken before the process is seriously upset.
Shewhart Controlcharts are referred to as Univariate Control charts because they observe asingle variable at a given time,Univariate charts consider individual qualitymeasurement sources and as a result they have several limitations. Applying univariate SPC results in the majority of thevariables collected on a process not being monitored. Furthermore, themonitored variables are not necessarily independent hence examining a limited groupof variables, one at a time, makes the identification and interpretation ofprocess malfunctions extremely difficult, and consequently the results of theanalysis may provide misleading information. Multivariate statistical process control methods (MSPC)address some of the limitations of univariate monitoring techniques byconsidering all the data simultaneously and extracting information on the’directionality’ of the process variations.For most industrial processes two ormore quality variables are important, and they can be highly correlated.
Forthese situations, multivariable SPC techniques can offer significant advantagesover the single-variable methods.