Statistical Process Control - Assignment Example

The term ‘statistical process control’ (SPC) refers to the usage of statistical tools with the intent of establishing control and monitoring a process to ensure that there neither is any lack of potentiality within the process nor any deviation (by any means) among the outputs (Oakland, 2003). SPC uses certain analytical tools and allows the investigator to see whether the process is working correctly as intended or not. By applying SPC one can maintain certain standards of a process in the absence of any inspection or regulation of every each and every step of the process and the final output.

It can be employed in any production unit (which generates material or data as outputs) where the process can thus remain independent from constant supervision. Although there can be variety in products and the required steps necessary for the production process, the SPC analysis remains uniform. SPC is advantageous over other methods of control particularly since it ensures that the production is not affected to any significant degree due to interruptions since it requires no direct inspection.

Only the "Out of control" process can be covered by SPC. An “in-control-process” does not always refer to an acceptable and desirable output. This conceptual misunderstanding often results in misleading applications of SPC. If not properly perceived, SPC can lead to loss of production.The Control chart is the fundamental tool for SPC. It is a graphical representation defined by the plot of measured data (at least 15 observations) on the chart. Using control charts the changes in the inherent process from the accumulated data can be identified.

This, in turn, determines the ‘special’ cause that affects the results. In a control chart the ‘central line’, denoted by a solid line reflects the average of the accumulated data points and passes through the middle of the graph. Control limits are the lines (a specified distance is maintained from the central line) in the control chart which are calculated by using statistical tools and indicate the behavior of the process i.e whether the process is ‘out of control’ or ‘in-control’.

There are two types of control limits- (i) the upper control limit (UCL) and (ii) the lower control limit (LCL). Recall that a Percentile is given by the division of pointed data in hundred equally divided groups. The nth percentile Pn is defined as- n% of the data points are on or below this value and (100-n) % data points are on or above this value. Thus, 90th percentile or P90 is a value such that 90% of the observed values are below this value and 10% are on or above the value. The 50th percentile represents the median of the data points.

(Montgomery, 1991) The control chart reflects that the process is well in control and working properly in its entire potential since there are no data points below the LCL. Additionally, the average data points are around the central line.