Statistical Process Control for Improving Business Performance in W L Gore & Associates - Research Paper Example

Application of Statistical Process Control (SPC) for improving Business Performance: Improving Manufacturing Quality in W. L. Gore & Associates Tableof Contents Table of Contents 2 Task 1 (Research Proposal) 3 1.1 Outline Specifications 3 1.3 Critical Review 4 2.1 Research Question or Hypothesis 5 1.4 Research Project Specification 6 1.5 Plan and Procedures 7 Task 2 (Final Research Project) 8 2.2 Research Investigation 8 2.3 Record and Collate Relevant Data 9 3.1 Findings 9 3.2 Interpretation of Results 12 3.3 Recommendations 13 Reference List 16 Task 1 (Research Proposal) 1.1 Outline Specifications In context to improvement of business performance in a globalized and competitive environment, companies are emphasizing on advanced analytics techniques and statistical mechanism in order to do predictive modelling on the basis of data generated through business process (Montgomery, 2005). According to Montgomery (2005), business performance of companies can be improved by deploying Statistical Process Control (SPC). On the basis of above theoretical argument, researcher has decided to conduct research in order to understand how implementation of Statistical Process Control (SPC) can help a specific manufacturing company to drive business performance by improving manufacturing quality. Problem statement: The intention of this research is to find out how implementation of Statistical Process Control (SPC) can help a specific manufacturing company to drive business performance by improving manufacturing quality Aim & Objectives- To understand how business performance is being positively affected through integration of Statistical Process Control (SPC). To understand how application of Statistical Process Control (SPC) can allow a specific manufacturing company to improve business performance through improvement in manufacturing quality. To understand the mathematical nature of functionality of Statistical Process Control (SPC) in business performance improvement. 1.2 Process of Research Project Selection Importance of statistical decision making is increasing in relation to business environment due to certain reasons, 1- companies need to measure quality in numeric terms so that they can improve quality in order to satisfy or even exceed customer expectations, 2- resources are getting limited; hence, accurate predictive modelling is needed in order to prevent wastage of resources owing to wrong decisions and 3- global data storage is increasing at an exponential rate and these data can be used for making key decisions (Montgomery, 2005). In such context, it has been found that statistical processes are being widely used by manufacturing companies in order to improve production quality. These theoretical arguments have influenced the researcher to select the research topic. While evaluating the research problem selection process, the researchers has found two qualifying criteria such as, 1- scarce amount of researches are available on the topic focusing on specific manufacturing company and 2- mathematical aspect of the research problem has been ignored by most of the previous researchers. These two research gaps have influenced the researcher to take up this study. 1.3 Critical Review According to Chakrabarty and Tan (2007), improving quality of the products, reducing faults in end products, improving productivity and increasing customer loyalty are considered as business performance indicators for manufacturing companies. In the manufacturing sector, quality means producing output without violating specification limits or coming to desired value termed as nominal value. Montgomery (2005) defined statistical process control (SPC) as a set of statistical techniques and mathematical algorithms that can be used to identify quality gaps and set quality targets. Chakrabarty and Tan (2007) argued that, in modern days, competitive advantage of companies depends on its ability to meet customer requirements by improving quality of output. Therefore, integration of SPC helps companies to meet customer requirements in responsive manner and achieve competitive edge over competitors through improvement of quality of output. According to Montgomery (2005), using SPC provides multiple of benefits, which can influence business performances such as, supporting decision making, measuring process variability, using the process variability to specify product requirements, eliminating variability in the process in order to lower frequency of errors in the end output. The literature regarding implication of statistical process control (SPC) in improving business performance is relatively new among research scholars. Importance of the topic has increased after Motorola improved its production quality through integration of SPC as part of Six Sigma process. For the next 20 years, plenty of research works have been done on understanding how SPC can help companies to enhance business performances. Notable fact is that, none can deny that the topic needs close integration of mathematics with management concepts. Unfortunately, for sake of simplicity, management scholars avoided mathematical analysis of the problem so that readers can easily understand their arguments. On contrary, due to lack of understanding in business and management concepts, mathematicians concentrated on developing mathematical theorem for the research problem and as a result, gap in the literature has been created. Bergman and Klefsjo (2003) conjecturally pointed out that no researcher has ever tried to balance mathematical aspects and business perspective of the research problem in equi-probable manner. As a result of that, gap in the literature has been created, which has forced the researcher to take up this study. Chakrabarty and Tan (2007) also pointed out that most of the researchers had tried to understand implication of SPC in business performance improvement by taking generalized approach for large set of organizations, while very few scholars have tried to direct their study at specific organizations. 2.1 Research Question or Hypothesis On the basis of the above discussion, following research questions can be derived. How business performance is being positively affected through integration of Statistical Process Control (SPC)? How application of Statistical Process Control (SPC) can allow W. L. Gore & Associates to improve business performance through improvement in manufacturing quality? What is the mathematical nature of functionality of Statistical Process Control (SPC) in business performance improvement? Resources for the research questions can be tabulated in the following manner. Table 1: Resource Tabulation Research Question Resources How business performance is being positively affected through integration of Statistical Process Control (SPC)? Chakrabarty and Tan (2007), Montgomery (2005), Bergman and Klefsjo (2003), Alwan (2000) How application of Statistical Process Control (SPC) can allow W. L. Gore & Associates to improve business performance through improvement in manufacturing quality? Franco, et al. (2003), Hoerl (2001), Alwan (2000), Drew, Glen and Leemis (2000), Sheikh (2003) What is the mathematical nature of functionality of Statistical Process Control (SPC) in business performance improvement? Evans, Drew and Leemis (2008), Rigdon and Basu (2000), Yilmaz and Chatterjee (2000) 1.4 Research Project Specification Historically, statistical process control (SPC) is rationalized as quality management tool that can be used mainly in manufacturing environment. Chakrabarty and Tan (2007) found that majority of research on statistical process control (SPC) are aimed at driving business performances in manufacturing environment. Therefore, the researcher has selected W. L. Gore & Associates as a manufacturing organization so as to understand how implementation of statistical process control (SPC) can help the company to improve business performances. Research questions and objectives can be summarized in the following manner. However, in real world scenario, different manufacturing companies use SPC in different manners in order to influence business performances. Therefore, the research should be done in case-wise manner (focusing on particular organization), rather than opting for a generalized approach to address the research problem. As a result, a gap has formed in the literature and has eventually forced the researcher to take up this study. Hence, the researcher has selected W. L. Gore & Associates as the sample manufacturing organization in order to conduct the research in a well-directed manner. From empirical perspective, Franco, et al. (2003) found that in recent years, W. L. Gore & Associates is facing stiff competition from other competitors in terms of product quality and manufacturing process optimization. According to Franco, et al. (2003), W L Gore & Associates is facing manufacturing quality issues in products like, fluoropolymer, versatile polymer and polytetrafluoroethylene (PTFE). Some common quality problems in manufacturing process of W L Gore & Associates are identified as high level of process variance, absence of quality standards, and high frequency of faulty products production, increasing manufacturing cost and lead time variance. Degradation of manufacturing quality is negatively affecting business performance of the company and as a result, business growth of the company is getting stagnated. As of now, as per knowledge of the researcher, no researcher has ever tried to analyze how application of SPC can help W. L. Gore & Associates to improve business performance through enhancement of manufacturing quality. With the given context, this research will shed light on the mentioned topic in order to determine how application of SPC can enable W. L. Gore & Associates to improve its manufacturing quality, thereby driving business performance. From the academic perspective, the research will try to develop mathematical model for functionality of SPC by using statistical concepts. 1.5 Plan and Procedures Bryman and Bell (2003) and Curwin and Slater (2008) pointed out that every researcher needs to follow certain research methodology in order to address research problems. In this paper, the research problem is more of theoretical in nature and data should be studied on the basis of their qualitative merit. Another problem is that for addressing the research problem in a quantitative manner, longitudinal study is needed; this is a costly as well as time taking process. Therefore, the researcher has decided to address the research problems through qualitative research. Analysis of existing literature and content analysis of secondary data available in online articles, company web sites, published case studies, annual reports, peer-reviewed journals and books will be used for addressing research problems (Creswell, 2009). Mathematical theorems will be used to formulate statistical framework of the research problem. As the research problem is broad and demands theoretical modelling, there is hardly any scope for primary research through questionnaire survey in this project. From ethical perspective, the researcher will not try to manipulate arguments presented in the secondary data and acknowledge each source to be used in research paper. It is evident from the discussion that significant amount of research scope exists for the researcher for the purpose of fulfilling the gap in the literature review, while addressing the research problems. Using qualitative research methodology will allow the researcher to address the research problem in a comprehensive manner. In the next section, the researcher will brief about the findings of the research that is conducted as per the chosen methodology. Table 2: Gantt Chart Months Month 1 Month 2 Month 3  Week Numbers  1 2 3 4 5 6 7 8 9 10 11 12 Activities     Research problem identification     Developing research proposal     Developing literature review     Completing qualitative analysis and refining research objectives on the basis of literature review Developing statistical model for the research problem     Analyzing the case study of the selected company Revising & final submission of prepared report Task 2 (Final Research Project) 2.2 Research Investigation W. L. Gore & Associates, Inc was established by and Wilbert (Bill) and Genevieve (Vieve) Gore and the company has headquarters located in Newark, Delaware (Gore, 2013). The company specializes in manufacturing both industrial and retail vinyl and polymer items like, polytetrafluoroethylene (PTFE), fluoropolymer and versatile polymer (Gore, 2013). Polymer products that are manufactured by W. L. Gore & Associates can be used in multipurpose manner, such as, manufacturing fabric laminates, signal transmission, medical implants, surf boats, water proof costumes and leisure items (Gore, 2013). Franco, et al. (2003) pointed out that quality of manufacturing process of the company has deteriorated; for instance, cost of production has increased without adding much to productivity, process variance has increased, frequency of faulty product production has taken a hike and variance in lead time is perturbing operations schedule of the company. In such context, the researcher has reviewed literature regarding the research problems and gone through cases studies published on the company. Therefore, findings of the research can be presented in the following manner. 2.3 Record and Collate Relevant Data Antony (2006) emphasized on Six Sigma as one of the most prominent Statistical Process Control (SPC) in case of manufacturing process. In case of Six Sigma process, maximum quality tolerance level is set at 3.4 parts per million or less that 4 faulty final products out of 1 million produced items (Hoerl, 2001). Implementing Six Sigma process helps manufacturing companies to improve performance quality of manufacturing in three different ways, such as, 1- identifying process variance in manufacturing and analyzing scope of quality improvement, 2- lowering process variances through process benchmarking and 3- implementing new process and standardizing the process in order to improve quality in a sustainable manner. It is expected that by implementing Six Sigma as SPC can aid W. L. Gore & Associates to reduce process variance by 99% and lower faulty products production to less than 1%. Hoerl and Snee (2003) used statistical control chart in order to identify variance in existing process and based on the control chart analysis, quality improvement plan can be designed. It has been found by the researcher that mathematicians use Least Absolute Value Regression (LAV) for the purpose of recognizing lag factors that can influence manufacturing quality improvement. In case of Least Absolute Value Regression (LAV) model, quality parameters are being set as explanatory variable; while key objective of statistical regression is set as smoothing sensitivity of residuals in process control chart. 3.1 Findings Hoerl and Snee (2003) argued that control process residuals may be autocorrelated and in such cases, SPC needs further adjustment. In non-mathematical term, it can be said that Least Absolute Value Regression (LAV) lowers summation of residuals or unexplained factors negatively, thereby affecting manufacturing process quality. LAV= ∑ (pi + ni); b+ mxi + pi + ni, i= 1,....n pi= positive manufacturing quality deviation from predetermined quality standard ni= negative manufacturing quality deviation from predetermined quality standard pi, ni >0 while m and b are unrestricted m= slope in regression line It is evident from above equation that LAV can be used as statistical process control by W. L. Gore & Associates in order to identify those manufacturing process factors, which need to be corrected with an immediate effect. In such context, Evans, Drew and Leemis (2008) upheld that in most of the cases, errors in manufacturing process show exponential distribution; therefore, it is better to employ Anderson–Darling Test Statistics, rather than linear parameters. In the later section, the researcher will use arguments of Evans, Drew and Leemis (2008) in order to assess the mathematical nature of functionality of Statistical Process Control (SPC). Consideration of research works of Hoerl and Snee (2003) reveals the fact that integration of Statistical Process Control (SPC) can improve business performance of companies in three distinct ways, such as, 1- improving manufacturing quality raises average productivity/employee for the company, 2- financial loss due to process variance gets lessened and 3- revenue generation opportunity for companies is improved as customers are satisfied with rise in quality of output. Franco, et al. (2003) had given detailed description of the manufacturing process of W. L. Gore & Associates. Manufacturing unit of the company works with four level partners, namely supply chain partners, product engineers, designing team and inventory management divisions. Local suppliers and third party vendors provide raw materials such as, polytetrafluoroethylene (PTFE), fluoropolymer and vinyl. These raw materials are stored in warehouses and cold storage is used for storing perishable items. On the other hand, product designing team conceptualize the product design and on the basis of their specification, raw materials are sourced. Computer-aided design (CAD) and 3D imaging are used for conceptualizing the product design as well as final outlook of the product (Franco, et al., 2003). In the later section, raw material is carried over to the manufacturing floor, following which the production engineers start their work. Logistics partners work closely with inventory management division so as to transport raw materials. Franco, et al. (2003) pointed out that engineering specifications for different product line varies significantly for W. L. Gore & Associates and such variance can be shown in the following manner. Figure 1: Process Variance in W L Gore & Associates (Franco, et al., 2003) It is evident from the widening gap in manufacturing process of W. L. Gore & Associates that the company is finding it difficult to reduce process variance. Due to such process variance, cost of production has appeared to rise, lead time variance has increased and faulty product production has also augmented. Examining the manufacturing process of the company can help the study to recognize problems. In case of W. L. Gore & Associates, automated robotic system arranges raw materials that are supplied and for most products, batch processes is used to develop standardized output. For example, composites base matrix embedded with ePTFE, Liquid crystal technology, cryogenic solutions and photovoltaic applications are employed for developing the final output (Franco, et al., 2003). Due to high process variance, the company is facing challenges in setting quality control chart for batch processes. This in turn results in absence of process benchmarking, which makes it difficult for the company to identify errors in the manufacturing process. For example, W. L. Gore & Associates uses 2 Layer protection mechanisms to devise waterproofing coat; but, the company has not established statistical limit for mixing of the 2 Layer protection coat. The quality of the end output is found to deteriorate due unequal mix of 2 protection layers (Franco, et al., 2003). As manufacturing quality of the company is found to be dropping, competitors like, Sympatex BHA Technologies or eVent, are posing threat to W. L. Gore & Associates by offering quality products at affordable price. Nonetheless, end products of W. L. Gore & Associates offer 100% waterproofing solution, while competitor’s product fail to offer such high level of water proofing solution. In such context, deployment of SPC can aid W. L. Gore & Associates to lower process variance up to its minimal level, thereby enabling the company to drive business performance through improvement of manufacturing process. 3.2 Interpretation of Results The mathematical nature of functionality of Statistical Process Control (SPC) in business performance improvement or improvement of manufacturing quality for W. L. Gore & Associates can be realized by using Kolmogorov–Smirnov (K–S) goodness-of-fit test. Let’s take, (cdf) ^F(x) = approximate cumulative distribution function for manufacturing quality in W. L. Gore & Associates ..... (Evans, Drew and Leemis, 2008) cdf Fn(x)= actual cumulative distribution function for manufacturing quality in W. L. Gore & Associates Now, normality of these two distribution function can be compared in order to determine fitness of quality. Fn(x) = frequency of manufacturing process time for developing final output ranging from X1……..Xn (observations) and these values will be less than or equal to mean manufacturing process time Fn(x) = I(x)/n, where I(x) = proportion of manufacturing process time less than mean manufacturing process time and n= total numbers of items being manufactured In such context, Dn= manufacturing quality gap or the Euclidean geometric difference between ^F(x) and Fn(x). Dn (Manufacturing quality gap) = supremum or greatest manufacturing process element {|Fn(x) - ^F(x)|} Dn+ (> manufacturing quality standard) = Maxi=1,....n{i/N- ^F(xi) Dn- (< manufacturing quality standard) = Maxi=1....n {^F(xi) – i-1/n} i= no. of manufacturing process encounter Based on the above statistical model, Goodness-of-Fit plot for manufacturing quality of W. L. Gore & Associates can be presented in the following manner. Figure 2: Goodness-of-Fit Plot for Manufacturing Quality (Source: Evans, Drew and Leemis, 2008, p. 1398) Through the use of Goodness-of-Fit Plot, it can be said that employing SPC can help W. L. Gore & Associates to determine specific quality gaps as well as to identify processes causing disruption in the manufacturing process. At a later phase, principal component analysis (PCA) or factor analysis can be applied to reduce dimensions of negative factors. As a result, manufacturing department of W. L. Gore & Associates gets a clear idea for improving manufacturing quality through implementation of quality control mechanism. Hence, it can be said that the above statistical model can be used by W. L. Gore & Associates to implement SPC in order to drive positive business performance through enhancement of manufacturing process. 3.3 Recommendations The researcher has gathered substantial insight from analyzing manufacturing process and value chain operation of W. L. Gore & Associates, besides developing statistical model for SPC integration in improving quality of manufacturing process. As a part of recommendations, value chain of W. L. Gore & Associates can be subdivided into following parts. Figure 3: Value Chain of W L Gore & Associates (Franco, et al., 2003) It is evident from the above diagram that entire value chain of the company works in two parts, such as, upstream positioning and downstream expansions. As of now, manufacturing team of W. L. Gore & Associates are involved only in upstream positioning like, converting PTFE to ePTFE membrane, polymer base manufacturing and laminating. Due to their absence in downstream expansion, such as, garment manufacturing or distribution, the company can neither control the entire value chain operation nor do they have any chance of assessing customer requirement in a realistic manner. In such context, they need to use SPC for identifying key success factors in the manufacturing process and integrating these factors in the manufacturing process so as to bring about positive business performance. It has been found by the researcher that the SPC functionality in business performance improvement can further be justified by using Kolmogorov–Smirnov or Anderson–Darling statistics. Companies like, W. L. Gore & Associates, can experiment with these statistical models in order to find optimal solution for using SPC in regards to manufacturing quality in a sustainable manner. Managerial Implication Recommendation mentioned in the paper can be incorporated by the management of W. L. Gore & Associates to improve its manufacturing process. Noticeable fact is that the statistical model can be utilized by product designing team of the company so as to set quality parameters. Even basic architecture of SPC can be designed by considering the recommendation. Areas for Further Consideration As further consideration, future researchers need to work with more robust statistical model in order to understand the way in which SPC can aid manufacturing companies to improve their process quality. The major limitation for the study is that the researcher has not tried to validate the developed statistical model through primary data analysis. In future context, researchers need to test validity of the statistical model proposed in the study through comparative analysis. Reference List Alwan, L., 2000. Statistical Process Control. Boston, MA: McGraw-Hill. Antony, J., 2006. Six sigma for service processes. Business Process Management Journal, 12(2), pp. 234-248. Bergman, B. and Klefsjo, B., 2003. Quality from customer needs to customer satisfaction. 2nd ed. Lund: Student literature. Bryman, A. and Bell, E., 2003. Business research methods. Oxford: Oxford University Press. Chakrabarty, A. and Tan, K., 2007. The current state of six sigma application in services. Managing Service Quality, 17(2), pp. 191-208. Creswell J. W., 2009. Research design: Qualitative, quantitative, and mixed methods approaches. 3rd ed. California: Sage Publications. Curwin, J. and Slater, R., 2008. Quantitative methods for business decisions. 6th ed. Andover: Cengage Learning EMEA. Drew, J. H., Glen, A. G. and Leemis, L. M., 2000. Computing the cumulative distribution function of the Kolmogorov–Smirnov statistic. Computational Statistics and Data Analysis, 34, pp. 1–15. Evans, D. L., Drew, J. H. and Leemis, L. M., 2008. The Distribution of the Kolmogorov–Smirnov, Cramer–von Mises, and Anderson–Darling Test Statistics for Exponential Populations with Estimated Parameters. Communications in Statistics—Simulation and Computation, 37, pp. 1396–1421. Franco, F. H., Song, Y. J., Tarazi, P. and Varma, G., 2003. The Rise and Rise of Gore-Tex. [pdf] INSEAD. Available at [Accessed 5th March 2014]. Gore., 2013. About Gore: Gore at a Glance. [online] Available at: [Accessed 5th March 2014]. Hoerl, R. W. and Snee, R. D., 2003. Leading six sigma. Englewood Cliffs, NJ: Prentice-Hall. Hoerl, R. W., 2001. Six sigma black belts: What do they need to know? Journal of Quality Technology, 33(4), pp. 391-435. Montgomery, D., 2005. Introduction to statistical quality control. 5th ed. New York: John Wiley. Rigdon, S. and Basu, A. P., 2000. Statistical methods for the reliability of repairable systems. New York: John Wiley & Sons. Sheikh, K., 2003. Manufacturing resource planning (MRP II): With an introduction to ERP, SCM, and CRM. New York, NY: McGraw-Hill. Yilmaz, M. R. and Chatterjee, S., 2000. Six sigma beyond manufacturing – a concept for robust management. IEEE Engineering Management Review, 28(4), pp. 73-80. Read More