The Issue of Quality Engineering - Research Paper Example

Introduction Quality engineering involves both quality assurance and quality control. Quality assurance deals with systematic activities that are implemented in a quality system in order to ensure that the required quality of a product is taken care of. In quality assurance there is systematic measurement comparing with certain set standards, the process are monitored with an associated feedback loop preventing the occurrence of error. On the other hand quality control tends to concentrate on process outputs. Under quality assurance there are several aspects that will be addressed including the quality of raw materials to be used, the assembling of the parts, products and the components being produced, the services that are related to the production process. The other aspects include the management process and the process of production and inspection process. Failure testing Failure testing is a very valuable process that is usually used on testing various consumer products. The process which is also known as stress testing stress testing can be described in mechanical terms as being as being operation of a product up to its failure point with the causes of failure being increased vibration, temperature and level of humidity. Through this test unanticipated weak points in a product are pointed out and the data is used in the improvement of engineering and manufacturing process. Statistical control The other process control used to improve quality is statistical control which has its basis both objective and subjective data. It is possible to investigate on quality of a product by stastical control as long as there is a common cause of variance or in case there is a special cause variance that may require tracking. It was recognized by Walter Shewart of Bell Telephone Laboratories on the process of making a product, it is possible to obtain data from specific parts so that the statistical variances can be put analysis. After the analysis it is possible to there can be implementation of control on the part through rework, scrap or there can be control implementation on the entire process which made the part with the aim of ensuring that the defect is eliminated before making more similar parts. Total quality management Under total quality management the quality of products will be dependant on all the constituent aspects. The process of total quality management can be well be understood by the contribution made by Juran under five concepts as follows. Spiral of progress According to Juran an organization the production and distribution of products is achieved through engagement of specialized activities that have there origin in various departments. The activities could be represented by a progressing pattern which was given a description of being spiral. The spiral gives a depiction of all the actions that are of importance of a product before the stage of introducing the product into the market. In this concept there is allocation to various departments the functions that are to be carried out. Sequence of breakthrough The sequence and breakthrough definition is given as a dynamic and decisive movement that is moving towards new and higher performance. In the breakthrough sequence the manner in which the activities are being executed brings about improved level of quality where the final result is good performance that would result into the launching of new products by the organization (Juran, 1964). It has also been noted that it is through breakthrough that quality leadership is achieved. Breakthrough also brings about solutions to many problems that are usually experienced at the field level in addition to improving the public image of an organization. Project by project approach Under project by product approach a quality improvement methodology is seen as one that need project by project during the implementation. The steering arm and the diagnostic arms are the two teams that were pointed out as being important by Juran according to (McConnell, 1987). Juran noted that the two teams had a clear vision of coming up with solutions to specific problems with a clear distinction being made between what was described as “putting patches” on problems and the elimination of problems after the problems have been searched and identified. The Juran Trilogy The Juran Trilogy was another major contribution which was made by Juran towards total quality management. This philosophy gives a systematic approach when carrying out his methodology in quality management. According to the trilogy the composition of management has three qualities each being interrelated this being: planning to achieve quality, improving the quality and quality control (Juran, 1986). Quality planning involves developing the process that enables the achieving of goals which have been established. Quality control sees to it that the gains which have would have been made are not lost in addition to avoiding increased waste. Through quality improvement there is lowering of the cost associated with poor quality in the existing process and also in addition learned lessons in bringing up innovative ways of achieving improved performance of a company. Vital few and trivial many The Pareto diagram was made important tool due to the fact that an emphasis was put on prioritization of problems by Juran. The Pareto diagram was based on the principal which was developed by an Italian economist who did research on how wealth was distributed among the people and he came to a conclusion that majority of the total wealth in the society was in the hands of few while the majority just controlled a small fraction (Ishikwa, 1982). In a simple was this fact can be stated as few factors tend to account to a very large fraction of the total. When Juran applied this principal in the industrial world it was possible to make a classification of the quality problem. Juran stated that most of the cost caused by poor quality have their origin from a very small number of causes called the vital few while the other portion are caused by the trivial many. Study methodology The study was with the aim of providing an insight into measuring the quality of components through use of a variety of measuring and sampling techniques. The study involve the interpretation of drawings and their tolerances so as to determine the actual number of rejects and the predicted number of rejects through use of statistical tables. There was also use of statistical process control (SPC) and Control charts. Percentage rejects The overall length and the slot length were measured for the first components with a total of 12 measurement made in each case. The overall length and step length was also measured for the second component where the step length was obtained by difference. The overall length in both cases was expected to range between 61.05 to 61.75 and all measurement outside the range were rejected. The step length was expected to range between 32 and 33 while the slot length was to be between 10.8 to 11.2. after the measurement were made the results were as in table 1 and table 2 and the same results represented diagrammatically as shown in figure 1. From the tables and figure it can be seen that out of the 12 components of type one 4 (33.3%) were rejected while 8 (66.3%) were accepted. In the case of component 2 out of the 15 samples 6 (40%) were rejected and 9 (60%) were accepted. Table 1 Table 2 Frequency Percent Accept 9 60.0 Reject 6 40.0 Total 15 100.0 Figure 1 Predicted number of rejects The mean and standard deviation were obtained and by use of statistical tables of area under normal distribution the number of rejects was calculated. The process used in as follows Predicting rejects in by overall length in component 1 Maximum Z-scores for overall length From figure 2 the mean and standard deviations have been given as 61.31 and 0.32. The maximum acceptable value is 61.75 Maximum Z-score = = =1.38 The area corresponding to Z=1.38 is 0.0.9162 This shows that 91.62% of the measurements are 61.75 Minimum Z-score = = =-0.81 The area corresponding to Z= -0.81 is 0.209 This shows that 20.9% of the recorded values falls below 61.05mm Therefore (91.62-20.9) % = 70.72% are acceptable as they fall in the allowable overall length Number of acceptable items = x12=8.4 Figure 2 Predicting rejects by slot length in component 1 Maximum Z-scores for slot length From figure 3 the mean and standard deviations have been given as 11.13 and 0.314. The maximum acceptable value is 11.2 Maximum Z-score = = =0.22 The area corresponding to Z=0.22 is 0.0.5871 This shows that 58.712% of the measurements are 11.2 Minimum Z-score = = = -3.34 The area corresponding to Z= -3.34 is 0.00042 This is an indication that none of the recorded values falls below 10.08 Therefore 58.71% are acceptable as they fall in the allowable slot length Number of acceptable items = x15=9 Figure 3 Predicting rejects by overall length in component 2 Maximum Z-scores for overall length From figure 4 the mean and standard deviations have been given as 61.38 and 0.11. The maximum acceptable value is 61.75 Maximum Z-score = = =3.36 The area corresponding to Z=1.38 is 0.0.9996 This shows that 99.96% of the measurements are 61.75 Minimum Z-score = = =-3.0 The area corresponding to Z= -3.0 is 0.00135 This shows that 0.14% of the recorded values falls below 61.05mm Therefore (99.96-0.14) % = 99.82 are acceptable as they fall in the allowable overall length Number of acceptable items = x15=15 Therefore all the measurement fall within the allowable tolerance Figure 4 Predicting rejects by step length in component 2 Maximum Z-scores for step length From figure 5 the mean and standard deviations have been given as 32.22 and 0.426. The maximum acceptable value is 33 Maximum Z-score = = =1.83 The area corresponding to Z=0.22 is 0.9664 This shows that 96.64% of the measurements are 33 Minimum Z-score = = = -0.52 The area corresponding to Z= -0.52 is 0.3015 This is an indication that none of the recorded values falls below 10.08 This shows that 30.15% of the recorded values falls below 32mm Therefore (96.64-30.15) % = 66.49 are acceptable as they fall in the allowable step length Number of acceptable items = x12=8 Figure 5 Comparison of actual and predicted percentage rejects The comparison of the predicted and the actual rejection is as given in table 3 and the graphical representation of the same being as in diagram 6. From the table it can be seen that in component one by overall length it was predicted that 29.28% would be rejected none of the component was rejected because of the overall length. For component one the predicted and the actual were almost the same with the actual percentage that was rejected being 40% and the predicted percentage being 41.29%. For component 2 in terms of overall length the actual percentage rejection was 8.3% while the predicted was 0%. The prediction for the actual rejection and predicted rejection in step length was equal with a value of 33.3%. These results are clearly represented diagrammatically in figure 6. Table 3 Actual Rejected Predicted O/L 0 29.28 Slot 40 41.29 O/L 8.3 0 Step 33.3 33.3 Figure 6 SPC sampling techniques -mean Using SPC sampling technique with 6 samples the result for the mean limit was as shown in figure 7. From the figure upper class limit of the mean value is gives as 61.49899; the average value is 61.29308 while the lower class limit is given as 61.08718. From the graph it can be seen that the process was out of control in sample 10 and in sample 12 where the mean values was below the lower class limit. Figure 7 SPC sampling techniques -mean The range limit was as shown in figure 8. From the figure upper class limit of the mean value is gives as 0.85380; the average value is 0.42608 while the lower class limit is given as 0. From the graph it can be seen that the process was out of control in sample 10 and in sample 12 where the range values was above the upper class limit. Figure 8 Probable tolerance Probable tolerance in length = = = = 0.25 Thus probable tolerance in overall length is 0.25 Probable tolerance in slot length = = = = 0.28 Thus probable tolerance in slot length is 0.28 Probable tolerance in step length = = = = 0.33 Thus probable tolerance in slot length is 0.33 Conclusion From the results it can be seen that in some of the tests the predicted values were almost the same as the actual values. On the other hand some of the predicted values had a very big difference compared to the actual values. This pattern could be attributed to some of the variables under investigation being far from normal distribution. It can also be concluded that the there was no results that were completely an expected. References Ishikwa, K (1982). Guide to quality control. White plains, NY: unipub-Kraus International. Juran.]. M. (1988). Juran on planning for quality. New York press. Juran.]. M. (1986). Management of quality(course materials). Wilton, CT: Juran Institute, Inc. McConnell, J. (1987). Analysis and control of variation. Australia: Delaware. Read More