Business & Management for engineers Essay Example
9Business and Management for Engineers
BUSINESS AND MANAGEMENT FOR ENGINEERS
Institution of Learning
The statistical process control depicts a systematic decision making instrument that allows an individual to view when the technique is not working and when working correctly. Variation is current in any procedure, choosing when variation is natural and when it wants rectification is the solution to quality control. Control charts indicate the variation in a calculation during the time phase that the procedure is viewed. In distinction, bell-curve kind charts, like process capability charts or histograms, indicate a snapshot or summary of the outcome. Control charts are a key tool of constant quality control. Control charts scrutinize processes to indicate how the procedure is performing and how the capabilities and process are affected by alterations to the process. That information is then applied to make quality advances. Control charts are also applied to determine the potential of the process. They can aid recognize assignable or special causes for aspects that hamper peak performance (Ab Rahman, Zain, Alias, and Nopiah, 2015).
Control charts indicate if a procedure is out of control or in control. They indicate the variance of the process output over time, like a measurement of temperature, width, or length. Control charts contrast that variance against lower and upper limits to view if it fits within the specific, expected, normal, and predictable variation levels. If so, the procedure is deemed in control and the variation among calculation is regarded normal random variation that is intrinsic in the procedure. If, however, the variation falls outside the bounds, or has run of non-usual points, the procedure is deemed out of control.
The statistical process control was undertaken as a method for application of statistical analysis to monitor, measure, and control processes. The chief statistical process control chief component is the application of control charting methods. The essential SPC assumption is that all procedures are subject to variation. That variation could be classified to two kinds including the assignable cause and chance cause variation. Among the SPC benefits include the capability to monitor a secure process and decide if alterations happen, due to aspects other than random variation. If assignable cause variation never happens, the statistical analysis assists in the source identification so it can be eradicated. Statistical process control also offers the capability to decide process ability, monitor processes, and identify if the procedure is operating as required, or if the procedure has altered and corrective action is needed (Oakland, 2007). Control chart information may be applied to decide the natural range of the procedure, and to contrast it with the precise tolerance range. If the natural range happens to be wider, then either the requirement range must be expanded, or perfections will be important to constrict the natural range.
Benefits linked with control charting emanate from both variable and attribute charts. If the control chart indicates that a procedure is in control, and within condition limits, it is often probable to eradicate costs associated to inspection. Control charts could be applied as an analytical tool show when alterations are necessary so as to foil the production of out of acceptance material. By the procedure, we mean the entire combination of producers, methods, suppliers, people, input materials, environment, and equipments that work concurrently to give output, and the customers who apply that output. The overall performance of the process relies on the communication between customer and supplier, the mode the process is designed and executed, and the mode it is managed and operated (Ab Rahman, Zain, Alias, and Nopiah, 2015). The rest of the procedure control system is positive only if it supplies either to improving the overall performance or maintaining the excellence level of the process.
One of the processes of the SPC is the dynamic process. A dynamic process denotes the process that is viewed across time. An SPC chart for indicating a dynamic process is usually denoted as a longitudinal or time-series SPC chart. Another process is the static process. A static process is viewed at a specific point in time. An SPC chart for indicating a static process is usually denoted as a cross-sectional SPC chart. A cross-sectional chart is vital in the comparison of different institutions.
One of the most significant departments of any organization is the sales department. The sales team requires having a lot of effort to attain success. The sales team ought to be very effective to the organization as it is dependent for the organization’s success. Sales performance monitoring is vital to maintain and finally meet their required objectives on their organization. Without the sales team plans and strategies to augment sales the organization will obviously fail. Sales department is accountable for anything that has association to business strategies in attaining more sales. From the service that the organization gives to promote and some publicity to your product organization is vital to have efficient strategies in sales performance monitoring. Sales performance monitoring is vital as the organization is all about sales. One of the ways used for observing the performance is the use of a balance scorecard (Oakland, 2007).
The balance scorecard is a management and a strategic planning system that is widely applied in industry and business, non-profit organizations, and governments globally to ally business activities to the strategy and vision of the agency, improve external and internal communications, and scrutinize organization’s performance against its strategic objectives. About 50 percent of the large firms in Europe have adopted the use of the balance scorecard. The balance scorecard has progressed from its initial application as a simple performance measurement structure to a full management and strategic planning system. A balanced scorecard offers a structure that not only offers performance calculations, but aids planners see what ought to be done and measured. It enhances executives to completely execute their strategies.
Calculation of control chart limits
This section prepares the provided data in a way that is useful in calculating the control limits
Mean for the first 10 subgroups
Mean for all the 25 subgroups
This data consists of k=25 (subgroups) and n = 4 (size of subgroups).This data is going to be used to construct an X-bar control chart. In the X-Bar chart, the center line is plotted at the level of the sample mean of the first10 sub group means. This mean is a natural estimate of the true mean and has been calculated as 349.7V
The control limits are estimates of , where the upper control limit is plotted at a level , and the lower control limit at . The is estimated from . To construct, there is a need to estimate σ. Using standard deviation to estimate it gives a biased result(Fryman 2001, p237). Another formula involving a correction factor d is used. Therefore;
, where r is the mean of the sub group ranges.
From the data, d2 = 2.0588, when n = 4, r = 2.6, and = 0.6436
Upper control limit (UCL) = = 349.7 + (3*0.6436) = 351.63
And the lower control limit (LCL) = = 349.7 — (3*0.6436) = 347.77
Fig 1: Control Chart
Discussion on the results
From the control chart plotted, it is observed that all the points are within the control limits. This indicates that the process is in control(Xie, Goh & Kuralmani 2002, p21). However, there is huge fluctuation with the limits. The manufacture should find out what causes this fluctuation and prevent it from reoccurring. Having all the points within the control limits does not guarantee that the process is always in control. There is a possibility of internal features of the chart that suggest instability and presence of other special causes. If there are points outside the control limits, it is advisable to look for special causes for these points. After finding the special causes, the offending points are removed and new limits calculated.
There are other criteria for out of control process, in addition to the points lying outside the control limits. For a stable process whose distribution of the mean is close to being symmetric, the probabilities of a point being within the center line are almost equal to half. A suggested criterion is that of looking for special causes if the following sequence of runs on one side of the mean occurs. (a) Having seven or more consecutive points on one side, (b) if more than ten out of eleven consecutive points are on the same side, (c) if twelve out of fourteen points are on one side and (d) more than fourteen out of seventeen points are on one side of the center line. The probabilities of these events occurring are 1.6%, 1.2%, 1.3% and 1.4% respectively. Because these probabilities are small, if any of the above events occurs, there will be a change in the process and possibly a shift in the process mean (May& Spanos2006, p182).
Other rules of detecting the shifts have been developed and are described by Nelson (Fryman 2001, p235). Assuming normal distribution, each of the eight patterns 0.5% probability that it will occur when the process is in control. The rules include (i) one point being outside the control limits, (ii) nine points in a row below or above the center line, (iii) six points in a row increasing or decreasing steadily, (iv) fourteen points in a row alternation up and down.
The trend indicated by (ii) above may suggest that the shift is resulting from causes such as tool wear, improvement of operator skills, or change in materials used. The fourth rule (iv) could suggest that there are too many adjustments done after every sample. The probability of getting a wrong signal from the four tests is about 1%. It is, therefore, advisable to look out for any abnormal behavior of the points.
According to Brussee (2004, p177). If it is economically desirable, it is advisable to carry out extra other tests to have early warning. This increases the probability of a false signal to close to 2%. The extra tests are based on dividing the region between the control limits into six zones. Each zone is one sigma wide, and there are three zones on each side of the center line. The zones are labeled A to C from the top while the bottom three are the mirror image. With these, there are two additional rules; (v) two out of three points in a row in or beyond zone A and (vi) four out of five points in a row in or beyond zone B. These additional rules are appropriate when there is more than one source of data.
The above-described rules have redundancies, as one rule may sometimes include the other. They should, therefore, not be followed slavishly. Any unusual pattern in the points is a sign of the need to look for a special cause.
Ab Rahman, M.N., Zain, R.M., Alias, A.M. and Nopiah, Z.M., 2015. Statistical process control. Maejo International Journal of Science and Technology, 9(2), pp.193-208.
Brussee, W. 2004. Statistics for Six Sigma made easy: Chapter 16 (p177) New York: McGraw-Hill.
Fryman, M. 2001. Quality and process improvement. (p235) New York: Autodesk.
May, G. S., & Spanos, C. J. 2006. Fundamentals of Semiconductor Manufacturing and Process Control. (p182) Hoboken: John Wiley & Sons.
Oakland, J.S., 2007. Statistical process control. Routledge.
Xie, M., Goh, T. N., & Kuralmani, V. 2002. Statistical Models and Control Charts for High-Quality Processes. (p21) Boston, MA: Springer US.
More Important Things