Statistical Process Control (Research Paper Sample)

Statistical Process Control
Student’s Name
Institutional Affiliation
Statistical Process Control
Introduction
Statistical Process Control (SPC) is a form of quality control that is implemented to monitor, control, and improve the various processes using statistical analysis tools. The industrial manufacturing processes may experience unpredictable and undesirable results rather than the preferred intrinsic variation. Thus, Statistical Process Control comes in handy in reducing the variations that are evident in a number of manufacturing processes, achieving optimal results and the best target value. It plays a crucial role in measuring and controlling the quality of a particular production process, ensuring that quality data that relates to the process measurements and the product is obtained. The control limits that are established in advance are influenced by the process’ capabilities while the specification limits of the production methods are determined by the needs of the client. The data, which is generated by the monitoring and control components of the statistical process control tool, enables managers to determine if the operations are operating within the required context. As a result, the data that falls within the control limits means that the firm’s operations are within the expected range. If the data falls out of the expected range, then it means that there exist product variations that emanate from the inefficient and unproductive processes. The SPC quality control technique is essential to the management as it enables them to make better cost control decisions, process improvement decisions, and to predict the output results and yields accurately. The purpose of this paper, therefore, is to provide an in-depth discussion of the basic types, variations, and characteristic of statistical process control techniques.
Two Fundamental Types of Inspection
The statistical process control method of quality control relies on sampling inspection in the collection of samples that are drawn from the products of the company in its manufacturing process. Sampling inspection is highly advantageous when compared to the traditional judgemental inspection techniques, given that it substantially reduces the amount of the inspection activity. In addition, sampling inspection enables the firm’s quality control process to provide feedback in the course of the production process. Consequently, it significantly reduces the likelihood of product anomalies, non-conformities, and the risk of product anomalies. It is because of the substantial reduction in the likelihood of anomalies that a company can be able to lower the rate of defects that are churned out of the firm’s production processes. In addition, sampling inspection ensures that the production staff can focus on the inspected sample by collecting the data and analysing it using statistical methods. It is applied to provide feedback about the manufacturing process and for decision-making purposes regarding the acceptance, rejection, or submission for rework of a given product lot.
There exist two broad sampling inspection methods that are used in the quality control tool of statistical process control. These are the single-stage sampling plan by attributes and the continuous sampling inspection technique. The single-stage sampling plan by attributes is utilized by a number of manufacturing firms to undertake sampling inspection of the products that are churned out of the manufacturing process. With regard to the sampling technique, a given number of pieces in the downstream production process should be drawn randomly and inspected. The certain number of pieces are drawn at any stage of the manufacturing process, depending on the following factors. First, the nature and quantity of the pieces selected is dependent on the batch size of the production run. Second, the number of the pieces that are drawn and inspected depends on the inspection level.
The other commonly used sampling inspection technique is the continuous sampling method. It is usually applied to the production processes that are of a continuous nature. Under this manufacturing flow, the products that are to be drawn and inspected are manufactured individually and continuously. Thus, the samples are not selected from the single batch that is churned out of the production process but from each product that is produced in a continuous single-product manufacturing process. During the course of choosing the samples to be statistically analysed for quality control purposes, the pieces are screened individually so as to identify their attributes. They are also checked for product and packaging conformity to correct the defectives. After a number of the prices have passed the inspection level and found to be fully satisfactory, a given amount of pieces are then checked randomly. In the random screening of the processed pieces, the discovered defective pieces are screened again for packaging conformity and product conformity.
The Significance of Natural and Assignable Causes of Variation
In the course of the quality inspection of a product, natural or assignable causes of variation can be found. It is the role of quality control to determine whether the variation is natural or assignable. The natural causes of the change result from natural patterns that are of usual, quantifiable, and historical nature. The evident changes in the products that are produced as a result of common causes include the application of inappropriate procedures, poor maintenance of machines, and poor design. Also, the natural causes of variation include factors such as the use of substandard material, a quality control error, a measurement error, normal wear and tear, and poor work conditions. Natural variation can also result from vibration in the company’s industrial processes, variability in settings, and ambient humidity and temperature that adversely affects the production process. Additionally, the absence of clearly defined standard operating procedures results in the company’s products having a high defect rate.
The assignable causes of variation are also referred to as the special causes of variation that are of an unusual, given that they have never been observed previously in the firm’s manufacturing processes. The assignable caused of variation are also of non-quantifiable nature, taking into account the fact that the company lacks historical data regarding the new, emergent, and non-anticipated phenomena. The assignable variation results from variations that are substantially not within the historical experience base of the company, thus, they are termed as inherently unpredictable. The occurrence of the particular variation results from some phenomena that have been previously rejected during the production processes. The causes of an assignable variation include poor adjustment of the manufacturing equipment, faulty controllers, operator absence, broken components, a poor raw material batch, and machine malfunction. In addition, other causes of the specific variation include computer crash, power surges, and system change-over failure.
The natural causes of variation are of high significance in quality control processes as it enables the firm to formulate actions that are aimed at reducing the noise in the system. It is also essential to understand the natural causes of variation since no specific action can be taken by the management in preventing the failures from occurring. The natural variation causes have severe implications on the firm’s product quality. They lead to the implementation of ad-hoc interventions by the management that may result in an increase in the frequency of defects and the level of variation. The assignable causes of variation in quality control processes are highly significant because they enable the management to adequately respond to the unexpected phenomena. The special-cause failures can be rectified by altering the process or component of the firm’s manufacturing process. Therefore, the assignable variations enable an organization to plan adequately for the emergent, unanticipated, and inherently unpredictable phenomena.
Task 3 LINK Excel.Sheet.12 C:\\Users\\Mike\\Documents\\#8183388455-SPC.xlsx "Descriptive Stat!R1C1:R16C2" \a \f 4 \h \* MERGEFORMAT 
Resistance (Ohms) Mean120Standard Error0.370479Median120Mode119Standard Deviation1.57181Sample Variance2.470588Kurtosis-0.39898Skewness0.613487Range5Minimum118Maximum123Sum2160Count18Confidence Level (95.0%)0.781643
The average resistance in Ohms of the resistors of various standard values is 120 Ohms, with a range of 5 and a standard deviation of 1.57181.
Normal probability curve
LCL=118.4282
UCL=121.5718
Mean=120
The data in the above case is assumed to be normally distributed with a mean of 120, a range of 5, and a standard deviation of 1.57181. The control process of the mass production process of the resistors of various standard values has a normal distribution, 95.0% of the population is captured by a normal curve at 1.57181 standard deviations from the mean. There is only a 5% chance of discovering a value beyond the 1.57181 standard deviations from the mean. A value of resistance in Ohms that is measured beyond 1.57181 standard deviations from the mean indicates that the mass production process has become unstable and more variable. The upper control limit for the mass production process is given as 121.5718 Ohms while the lower control unit is given as 118.4282 Ohms. The process mean is given as 120 Ohms and any deviations that exceed 1.57181 standard deviations from this means indicate the instability and variability of the mass production system.
The selection and grouping ...