APPLICATION OF SPC TOOL FOR FINDING VARIATION IN THE PROCESS OUTPUT A CASE STUDY. Manav Ravchna International University, Faridabad (India)

Transcription

1 International Journal of Industrial Engineering Research and Development (IJIERD), ISSN (Print) ISSN (Online) Volume 2 Issue 1, May October (2011), pp IAEME, IJIERD International Journal of Industrial Engineering Research and Development (IJIERD), ISSN 0976 I A E M E APPLICATION OF SPC TOOL FOR FINDING VARIATION IN THE PROCESS OUTPUT A CASE STUDY ABSTRACT 1 Sahil Sardana, 2 Rajender Kumar, 3 Manjinder Bajwa, 4 Piyush Gulati 1 Student, M.Tech., YMCA Univ. of Technology, Faridabad (India) 2 Asst. Prof., Department of Mechanical Engineering, Manav Ravchna International University, Faridabad (India) 3,4 Asst. Prof., Department of Mechanical Engineering, Lovely Professional University, Jalandhar, Punjab (India) #Corresponding Author, address: rajender629@yahoo.com, rajender.fet@mriu.edu.in In this context, in order to develop, and even to survive, an enterprise must be competitive. In the past few years, several deployments are derived from SPC techniques in order to quantify and analyze project risk issues. The cost of failure of a component is not only the cost of replacement of component but it may involve many other expenses occurred due to the delays in work. For a customer it may be only monetary loses but for a company, it effects the reputation also. In this research work, an attempt has been made to study the manufacturing processes to reduce the defect rate by using the Statistical process Control (SPC) techniques, thus the various possible cause of failure and their effects along with the prevention are discussed. The present work deals with the study of different products i.e. shocker seals in SIGMA FREUNDENBERG NOK. The basic requirements of the manufacturing processes are studied. The Analysis of Variance (ANOVA) through X bar & R Charts are used to analyze the results to determine how much variation each quality-influencing factor has contributed in the rejection of the product. KEY WORDS Management Policies, Statistical Process Control, Statistical Thinking, Control Charts, Quality Assurance. 1.0 INTRODUCTION: QUALITY & COMPETITIVENESS Globalization of markets has brought them all closer together and resulted in progressively greater competition. The idea of local markets and protected markets has undergone a sea change. In this context, in order to develop, and even to survive, an enterprise must be competitive. This implies that the customer must be offered a package 46

2 (a product and/or a service) of high quality and at the lower possible price. These two essential requirements: that may seem to be contradictory, actually go together. It would therefore, not be unreasonable to say that to ensure quality is to: remain in business, retain employees, achieve technical progress, reach and develop new markets for the products, and finally make profits from sale. The literal translation for SPC is as under: Statistical : Adjective Process : Process Control : Management (Mastery) SPC may be defined as an element of Quality Assurance in production and a tool for continuous improvement. With all that the term Quality Assurance implies, the term production taken in its wider sense: production of goods, provision of services and performances. SPC is the application of statistical methods to the monitoring and control of a process to ensure that it operates at its full potential to produce conforming product. Under SPC, a process behaves predictably to produce as much conforming product as possible with the least possible waste. While SPC has been applied most frequently to controlling manufacturing lines, it applies equally well to any process with a measurable output. Key tools in SPC are control charts, a focus on continuous improvement and designed experiments. Walter A. Shewhart introduced SPC in the early 1920 s. W. Edwards Deming later applied SPC methods in the United States during World War II, thereby successfully improving quality in the manufacture of munitions and other strategically important products. 1.1 BASIC IDEA IN SPC SPC is based on three principal realities that are interrelated: it is the manufacturing process which determines whether a product is standard or substandard. Processes are schizophrencis : their behaviour varies over a period of time. Processes are entropic : they have a natural tendency to become disorganized and deteriorated in time. For each of these realities SPC suggests a method and/or a solution. In all cases, however, it provides a single tool, the well-known Control Chart. (a) The First Reality: It is the process that ensures Quality When go further for detail about the main reasons for defect, finds the main Causes Man, Machine, Material, Method and Money (Generally known as 5M s). There is no clear understanding weather the final results is acceptable or not. Merely controlling the final product (by inspecting it) without acting on the process will not really help in improving quality. Inspection should therefore, be replaced by control. This implies acting on the cause of the defects (the process), instead of acting on the effects (defective products). This is the main idea and the starting point for SPC. 47

3 By controlling the process, and since the process is upstream of then finished product, the cost associated with tests, inspection, rejections, and guarantee services, etc. are greatly reduced. This is a direct consequence and the attractive aspect of SPC: reduction in costs. (b) The Second Reality: Processes and Schizophrenic Often the processes constantly fluctuate from standard. When they appear to be stable, it is because of the period of observations is too short or because of the measuring precision is inadequate. Such fluctuations are caused by the multiplicity and variability of the different parameters affecting a process: the raw material may vary from one lot to another, the temperature and humidity may vary from day to day, certain adjustments may vary from one team to another etc. SPC helps in analyzing such fluctuations, understanding the, and finally in reducing them to keep them with in limits that are compatible with specifications or objectives. (c) The Third Reality: Process are Entropic The natural tendency of processes is to become disorganized and degraded over a period. This involves a change for the worse. The process itself is the main dependent variable for the extent of this drift. SPC enables to follow-up this degradation, to quantify it, and to act effectively on its causes. What is more, SPC enables to reverse the tendency. It means, Instead of continuous degradation it will also helps in continuous improvement. 1.2 Aspects of SPC SPC mainly consisting four major aspects, which are described below: The Idea: Since SPC is the process that causes a product to have, or not to have, a defect. It is the process that must be controlled instead of the product. The Approach: The objective of SPC approach is to control the parameters affecting the process in effective manner. The approach involves a series of successive stages from motivating the personnel to the use of control charts to maintain a check the characteristics of the product and the corresponding parameters of the process. The Tool: Control Chart is the simplest tool for controlling the process variation. The simplicity of a control chart doesn t affect its effectiveness. This is precisely where the effectiveness of SPC lies: in its being simple to use and in the effectiveness of its results. The Concept: Quality improvement is a continuous and endless process. It is always possible and always worthwhile to make such progress. Zero Defect is neither sufficient nor a limit. G. Taguchi has shown that even when a characteristic is within the laid-down limits of tolerance there is loss of society when the characteristics are nearer the target value. 1.3 Procedure for using SPC Walter A. Shewart(1931), suggested the keys for any process improvement program through SPC is the PDSA cycle shown in Fig

4 PLAN: Identify the problem and the possible causes. The QC tools described in this manual can help organizations identify problems and possible causes, and to prioritize corrective actions. DO: Make changes designed to correct or improve the situation. STUDY: Study the effect of these changes on the situation. This study shows the effects of changes on a process over time. Evaluate the results and then replicate the change or abandon it and try something different. ACT: If the result is successful, standardize the changes and then work on further improvements or the next prioritized problem. If the outcome is not yet successful, look for other ways to change the process or identify different causes for the problem. 1.3 Introduction to Control Charts Fig. 1 PDSA cycle Control Charts is an essential tool of continuous quality control. They monitor the processes to show how the process is performing and how the process and capabilities are affected by changes to the process. This information is then used to make quality improvements. They can help to identify special or assignable cause for factors that impede peak performance. It helps in finding the variance in the output of a process over time, such as a measurement of width, length or temperature etc. It compares this variance against upper and lower limits to see if it fits within the expected, specific, predictable and normal variation levels. If so, the process is considered in control and the variance between measurements is considered normal random variation that is inherent in the process. If, 49

5 however, the variance falls outside the limits, or has a run of non-natural points, the process is considered out of control. The control charts like X bar & Range Charts are the statistical tool used to distinguish between variations in a process resulting from common causes and variation resulting from special causes. It presents a graphic display of process stability or instability over time. X Chart: In X chart, means of small samples (3-5) are taken at regular intervals, plotted on a chart, and compared against two limits. The limits are known as upper control limit (UCL) and lower control limit (LCL). LCL = X - A 2 R & UCL = X + A 2 R (Where n is the no. of observations) Where, X is the target mean and factor A 2 depends on sample size? The process is assumed to be out of control when the sample average falls beyond these limits. Now for finding the average of subgroups used following equation. (Where k is the number of subgroups) Upper control limit: Lower control limit: Range (R) Charts: In these charts, the sample ranges are plotted in order to control the variability of a variable. The centreline of the R chart is known as average range. The range of a sample is simply the difference between the largest and smallest observation. If R 1, R 2,..., R k, be the range of k samples, then the average range (R bar) is given by:- The upper and lower control limits of R chart are: Upper control limit: Lower control limit: Where, factors D 3 and D 4 depend only on sample size (n) 50

6 2.0 LITERATURE REVIEW From the past literature survey, it is evident that some research on SPC have been carried by previous researchers but still a lot of applied research in this field is required so as to explore the utilization of SPC technique in the area of production, manufacturing and design. Shewhart (1931), in his first book on quality control, described to select the criterion (the three sigma limits) that we use today. A control chart predicts that, in the absence of assignable causes, the process will operate as a random system and produce the present level of quality in the future. If that level of quality is not satisfactory, a fundamental change in the process is required. Juran (1964), studied how problems can be solved successfully. In a major work, Managerial Breakthrough, Juran explained the essential elements required for problem solving and improvement. It is interesting to note that Juran s teaching was picked up by the Japanese and returned to the United States as Quality Control Circles. Ishikawa (1985), known as the father of Quality Control Circles, added the cause-andeffect chart as an aid to brainstorming, but all the rest of the methods were the same as those recommended by Juran. The popularity of these problem-solving tools and the ease of their use caused many to adopt them and neglect tools such as the quality control chart, which had the stigma of the word statistics. The non-manufacturing sectors such as government, and service seemed especially to shy away from quality control charts. This was not the intent of Dr. Ishikawa. He looked at three levels of what he called Utilization of Statistical Methods. In order of difficulty, these are as follows: Elementary Statistical Method (also known as the Seven Tools of Quality ) Intermediate Statistical Tools Advanced Statistical Methods (using computers concurrently) (Ishikawa 1985) Deming (1992), in his book, Out of crisis had given more than 100 suggestions for applications of control charts. These cover such diverse areas as sales personnel, motor freight, administrative applications of a manufacturer, health care including hospitals, airlines, Bureau of the Census, hotels, restaurants, city transit system, railways, telephone companies, department stores, banks, electric utilities, and municipal services. Kane (1986), carried out a study on capability indices and examined uses of capability indices along with their estimation procedure. When properly applied, statistical tools are an effective way for improving process quality. Yung(1996), suggested an integrated model for manufacturing process improvement in PCB manufacturing. In this, the statistical process control and seven basic tools have been integrated for valuable and effecting benefits in process improvement. Klenz (1999), discussed the method of data warehousing to give data marts and info marts that can be used for decision making by yielding meaningful information. Control charts are part of his database. Prajapati and Mahapatra (2007), proposed a joint X and R chart to monitor the process mean and variance simultaneously. They have suggested a very simple and 51

7 effective design of proposed joint X and R chart to monitor the process mean and standard deviation. Sultana et al. (2009), intended to combine the Hourly Data System (HDS) and Statistical Process Control (SPC) practices to improve manufacturing performances in manufacturing companies. The focus of their work is to find out the frequencies and time duration of machine breakdowns as well as the major causes of breakdowns affecting productivity. 3.0 INTRODUCTION OF INDUSTRY & PRODUCT In August' 2000 Delhi based SIGMA, FREUNDENBERG of Germany & NOK of Japan formed a joint venture company, named Sigma Freudenberg NOK Pvt. Ltd. (SFN) to manufacture wide range of rubber seals in India. This joint venture in India is will bring the latest technology in design, material, application engineering and production to India to support their Indian customers with state of the art seals, giving them a competitive edge in a more and more competitive environment. Production started in November 2001 and will continue to increase over the next years. Together with the trading business, Sigma Freudenberg NOK (SFN) will be able to offer a unique range of seals for every kind of application in the Engineering Industry. It has got quality certifications like ISO/TS 16949:2002; ISO 14001; OHSAS & ISO Being the unchallenged market leader in the sealing technology, the Freudenberg and NOK, group of companies together with their Indian partner, Sigma is offering its customers the widest and latest range of sealing solutions for various applications in the automotive and other related fields. Its product range offers following class products: Oil Seals Radial Shaft Seals Shock Absorber/Rod Seals Hydraulic Seals Shaft Seals Valve Stem Seals 4.0 ANALYSIS FOR FINDING VARIANCE IN OUTPUT: APPLICATION OF CONTROL CHARTS In the present work, 100 samples are inspected by the Quality Inspectors. Each & every sample having a subgroup of 4 no. of Shocker Seals. All the inspected value for the samples is given in Appendix-1. Calculations for Sample No.1 are: X = X 1 + X 2 + X 3 + X 4 4 X = ( )/4 = Similarly finds the calculations for 400 Samples & the values are shown in Appendix-1. Now, finding the average value of X from 400 samples is as follows: 52

8 k = i= 0 X /100 = /400 = mm For the subgroup of 4:- A 2 =0.738, D 4 = 2.28, D 3 = 0 Upper control limit (UCL X )= *0.0911= mm Lower control limit (LCL X )= *0.0911=61.94 mm Figure 2: X Chart for showing variation in product dimension with in limits Now finding the Range: R Bar =9.11/100 = mm (where k = 100) Upper control limit = 2.28* =0.21mm Lower control limit = 0*0.0911= 0 53

9 UCL r LCL r RANGE CONCLUSION & FUTURE SCOPE Figure 3: Range Chart The contribution of individual quality influencing factors is the deciding key of the control to be enforced on the product design. The confirmation experiments also indicate a clear picture of variation of Shocker seal product. Figure 2 and Figure 3 shows that the products are vary in the control limits. So, the rejection rate of the product is zero. Hence the Zero Defect is achieved with in the plant. In future, SPC tools are used to eliminate cumbersome, low-value tasks. As in this research work only control charts for variable data i.e. X bar and Range charts are used, other control charts like CUSUM, EWMA, etc can also be used. Also control charts for attribute data like p-chart, np-chart and c-chart can be used. Moreover large-scale production, with the aid of new software SAS (statistical analytical software) process control can be done more easily and accurately. Other Methods of assuring quality are JIT, TPM, Taguchi, and Lean Manufacturing etc. REFFERENCES 1. Carleton, A. D. (1999), Improving the Software Process with SPC. In ASQ (Ed.), 53rd Annual Quality Congress Proceedings, Milwaukee, WI: American Society for Quality 2. Chase, N. (1999), The Uses and Abuses of SPC. Quality 38(5): Deming,W. E. (1992), Out of the Crisis. Cambridge, MA: Massachusetts Institute of Technology, Center for Advanced Engineering Study. 54

10 4. Deming, W. E. (1994) The New Economics for Industry, Government, Education (2nd ed.) Cambridge, MA: Massachusetts Institute of Technology, Centre for Advanced Engineering Study. 5. Ishikawa, K. (1985), What Is Total Quality Control? The Japanese Way (D. J. Lu, trans.). Englewood Cliffs, NJ: Prentice-Hall, Inc. (Original work published 1981). 6. Jones, G. T Strategic Planning: Unleashing the Mystery Force, In ASQ, ed., 53rd Annual Quality Congress Proceedings: Milwaukee, WI: American Society for Quality. 7. Juran, J. M. (1964), Managerial Breakthrough, New York: McGraw-Hill. 8. Klenz, B. W. (1999), The Quality Data Warehouse: Serving the Analytical Needs of the Manufacturing Enterprise, In ASQ, ed., 53rd Annual Quality Congress Proceedings: Milwaukee, WI: American Society for Quality. 9. Prajapati, D. R. and Mahapatra, P. B. (2005), Control Charts for Controlling the Process Mean: A Literature Review, Journal of Udyog Pragati (NITIE), Vol. 29, No. 2, April-June, 2005, pp Prajapati, D. R. and Mahapatra, P. B. (2006), An Efficient Design Approach for X and R Control Charts, Journal of Institution of Engineers, IE (I)-PR, March 2006, Vol. 86, pp Shewhart, W. A. (1939), Statistical Method from the Viewpoint of Quality Control. Washington, DC: The Graduate School, The Department of Agriculture. 12. Sultana F., Razive N. and Azeem A.(2009 ), Journal of Mechanical Engineering Vol. ME 40, No. 1, June 2009 Transaction of the Mech. Eng. Div., The Institution of Engineers, Bangladesh 13. Woodall, W. H., Spitzner, D. J., Montgomery, D. C., and Gupta, S. (2004), "Using control charts to monitor process and product quality profiles." Journal of Quality Technology, 36(3), pb Zaidi, A. (2003), SPC (Concepts Methodologies, and Tools), Eastern Economy Edition, Prentice Hall India, Delhi. APPENDIX - A S. NO. X 1 X 2 X 3 X 4 R

11

12

13 Total