Using PDCA Cycle and QC Tools to Reduce Defects in a Manufacturing Industry

Using PDCA cycle in correlation with QC tools to Reduce Defects in a Manufacturing Industry

Abstract

This paper intends to use the theoretical and practical application of quality improvement using Plan-Do-Check-Act (PDCA) Cycle in the production line of a manufacturing environment. The paper mainly focuses on finding out the rate of rejections which affects the production line. We also try to analyze and find out the major causes which are the reasons for such rejections using quality tools. Successful implementation of the recommendations of this paper can significantly improve the manufacturing performance of a manufacturing environment

Keywords

Statistical Process Control (SPC), Quality Tools, PDCA Cycle.

 1. Introduction

Total Quality Management is defined as “The management that consists of organization-wide efforts to install and make permanent a climate where employees continuously improve their ability to provide on demand products and services that customers will find of particular value”(Ciampa, 1992). The word ‘total’ emphasizes that all departments in a company are responsible for collective improvement of their operations and ‘management’ emphasizes that the executive of the companies are strictly managing the quality through different methods. Statistical Process Control (SPC) is one of the main tools of Total Quality Management. SPC is somewhat a similar process to TQM but not as a whole. TQM involves the whole lot of employees working for continuous improvement of the work and its culture. Whereas SPC is a controlled process which is only concerned with improvement of a specific process. SPC has these days become one of the most popular and widespread solution techniques in industries.

Shop floor optimization these days is a major headache for most of the manufacturing industries. Problems is the shop floor can disrupt the assembly and production of the companies leading to immeasurable loses. We can use Hourly Data System (HDS) to help us visualize and view the real time production flow of any manufacturing company. HDS even helps in quality control by providing the basic data. Mostly SPC processes have been successful in solving problems in manufacturing production lines. This paper will deal with a common problem in manufacturing plants, rejection of lots of equipment parts which leads to disruption of production lines along with wastage of time and resources.

1.1   Motivation

With the growing demand for better quality and fast production of components the process needs to be properly optimized and the root causes where the quality is affected needs to be found out. On the current world market both manufacturers and consumers require guarantees for the quality of products and services. One of the ways to ensure that the required quality is obtained at appropriate cost and time is by applying quality control in the organization. The aim of these companies is the same i.e. to produce high quality and reliable products, meet customer expectation, fulfill ISO conditions and compete with the market.

1.2   Objective

Among various applied quality control techniques in the companies, SPC is the most popular applied techniques. Control chart, check sheet, and cause and effect diagram were used in these companies to analyses and interpret the data related to product quality.  SPC is not only easy and simple to be used, it provides a collection of powerful problem-solving techniques to achieve process stability and reduce variability, and can be used in any process (Montgomery, 2005). Consistent with previous research (Srikaeo et al, 2005; Mason and Antony, 2000; Antony and Taner, 2003; Xie and Goh, 1999), control chart is the most widely used SPC techniques in industries specially to monitor production process. So, with the help of using appropriate Quality tools the defects in the manufacturing industry can be reduced with a big difference value. In this study our aim is to find out the root causes due to which these defects occur and try and find them out by the tools.

1.3   Flow Process

Table 1. Correlation of PDCA with QC tools

We use PDCA cycle in conjucture with QC tools. We use Check sheet and Pareto Chart for the problem identification. Using Histogram and Cause and Effect diagram we work on problem analysis. Control charts and process capability reports are then used to provide plausible improvements inputs.

2. Literature review

In recent times, there has been a lot of concern on the quality of life. Statistical Process Control (SPC) is a statistical approach for assisting operators, supervisors and managers to manage quality and to eliminate special causes of variability in a process (Oakland, 2003). The initial role of SPC is to prevent rather than identify product or process deterioration, but Xie and Goh (1999) suggest for its new role to actively identifying opportunities for process improvement. The main tools in SPC are control charts. This has especially been evident in the workplace where the atmosphere, culture and ergonomics of the workplace have come into light. Noise matter has become one of the tops most priorities because of it long term effects on employees. As a considerable source of noise in an industrial environment, the Induction motor is one such element which needs to be worked on. Total Quality Management is defined as “the management that consists of organization-wide efforts to install and make permanent a climate where employees continuously improve their ability to provide on demand products and services that customers will find of particular value”(Ciampa, 1992). The study of sources and elimination of inductor motor noise is a very broad field where there is still lot of work put in. It is important that this motor is technically improved in order to just supply more mechanical power but not increase the noise output. Although hundreds of articles for practitioners have discussed specifics of TQM, and most experts agree that properly implemented quality systems improve organizational performance (Garvin, 1987). In an Induction motor, one part of the input power is dissipated as heat, another part is spent by the ventilating system and a third part, smaller than the previous ones, is lost as sound noise. Even though this kind of power loss as sound is very small, can be very significant if the motor noise level goes beyond limits described in standards.

“One of the solutions everyone came up with was to enclose the motor with a material which was sound absorbent, but it was found to be not feasible economically and it might modify the thermal characteristics of the motor” (Nau, 2006).  This study recognizes the best way to analyze and reduce the noise output from the motor is right at the start at its design stage. It is best to analyze, find the causes of noises and optimize the design, solving the problem right in the design stage.

Statistical Process Control has been available for many years, (Shewhart 1920’s) and has been adopted by large manufacturing companies. “SPC is a method which gives confidence that components produced are within tolerance, without having to measure every component. It is associated with the theme of controlling the process NOT the product and is a form of feed forward Control” (Lucas Manufacturing Systems Engineering Handbook, 1988). Some basic tools of SPC is the average and range control chart, for variables and attribute data and p, np, c, u charts, machine capability etc. Resent developments of charts include moving average, moving range and short run SPC (Nau, 2006). De Magalhães et al. presented a very clear overview of these techniques, publishing a key paper in this field (De Magalhães et al, 2009). Donald S. Bruce published a similar paper in which he optimized the supply chain of Cooper Tire and Rubber Company by optimizing the supply chain with low cost, high quality raw materials using SPC (Bruce, 2013). Similarly, Md. Sourove Akther Momin, Md. Golam Kader and Md. Mahbubur Rahaman published a paper in 2016 at International Journal of Mechanical Engineering and Automation where they studied and analyzed the reason for rejections in a pipe manufacturing industry using SPC (Momin, 2016).

There are several papers which deal with reduction of noises in electrical motors including both Induction and DC motors. One such study was conducted by P. Pfliegel, F. Augusztinovicz and J. Granát, where they study and analyzed the cause of noise in small DC motors but were unsuccessful in providing concrete results. According to Sebastião Nau, who published an article in 2000 about the Causes and Solutions of acoustic noise in Induction Motors, the best line of action is to analyze and optimize the motor right in the design phase of it. In relation to the three sources of noise in an electric motor one can say that for a sinusoidal supply and constant speed, airflow noise is the more important cause for 2 and 4 poles motors and magnetic noise becomes the largest source for motor with more than 6 poles (Nau, 2006).

Parkash et al conducted their work on sarcastic process management, during which they delineate SPC, different types of method capabilities, performance index, 5M’s, completely different tools for SPC, different monitoring tools like CUSUM charts, EWMA charts and implementation of SPC (Prakash et al., 2009). R. Srinivasu, G. Satyanarayana Reddy and S.R. Rikkula conducted their work on utility of internal control tools and SPC supported up the productivity and quality in trade, during which they focused on the standard assurance victimization the standard control tools. They additionally delineate however the method could be helpful each for the makers and customers (Srinivasu et al., 2011). J.-H. Yang and M.-S. Yang conducted their work on management chart pattern system supported applied mathematics correlation coefficient methodology, during which they tried to explain that statistical coefficient of correlation approach is also a very straightforward tool to acknowledge these unnatural management chart patterns effectively. This method is additionally Associate in Nursing effective methodology for the management chart pattern while not a tedious coaching method (Yang et al., 2005).

“Accurately identifying noise and vibration sources for small motors is challenging due to the small physical dimensions of these machines and their relatively wide frequency range” (Cho, 2017). According to Yong Thung Cho, the range of frequencies for the motor shown in the present work was 84 Hz to 7192 Hz, and the highest frequency that avoided spatial aliasing of the measurements was 8575 Hz. Overall, major sources of motor noise and vibration were electro-magnetic forces, internal resonance, and motor housing resonance. Unbalanced forces on the rotor, top cooling holes on the motor housing, and the switching of brushes were also dominant sources of noise at 84 Hz, 252 Hz, and 508 Hz, respectively, which is clearly shown through the reconstructed particle velocity of source surfaces via measurement pressure (Cho, 2017).

This paper was possible thanks to a similar work conducted on machine breakdowns in a Cigarette production Plan (Sultana et al., 2009).

3. Theory

For analysis the machines, the volume of Titanium HSS Drill Bits packaging unit (CPD) was taken and analyzed. There are many different types of taper drill bits produced by the company as per the demands. The drill bits size and shape vary according to the nature and angle of the surface to be drilled. So daily production volume is recorded both machine wise and type wise.

Figure 1. A conventional Titanium HSS Drill Bit

The company also manufactures the following class products:

• HSS Drill Bits
• Cobalt HSS Bits
• Reduced Shank HSS Bits
• Diamond (Tile) Bit
• Black Oxide HSS Bits

We consider only the Titanium Drill Bits drills for the vast size of the data and to avoid complications.

The CPD data was obtained from a student colleague who interns at the company. This information square measure then aforethought in Minitab, which may be an applied mathematics code for production trend analysis. Root causes of failure for production information which is out of management limit square measure then searched for from HDS, Maintenance Schedule, Communication Log Book, and Electrical Log Book so the findings are analyzed. After it recommends square measure given supported findings. The sequence of the study is given in the form of a flow chart below.

3.1 Description of Methodology

Quality tools are the main basis of function for this case study. We are going to use the Plan-Do-Check-Act cycle in correlation with QC tools to identify the rejections when producing different kinds of taper shank drills and what are the rejection causes and reasons. The quality tools are very important and are used in many of the companies to solve quality issues. Ishiwaka, who was the founder of such tools has claimed that it is possible to solve 95% of quality problems with these tools. The following are the 7 QC tools:

1. Check-Sheet
2. Pareto Diagram
3. Cause & Effect Diagram
4. Scatter Diagram
5. Histogram
6. Run Chart
7. Control Chart

3.2 Check Sheet

Check Sheets are one of the most important type of quality tools used widely. The main purpose of a Check sheet is to collect and analyze real time data at whatever location the data is generated. This data from check sheet is usually simple and is used in conjunction with pareto chart and histograms. We have a data of rejection parts and their defects as shown in Table 2.

3.3 Pareto Diagram

A pareto Chart is solely a frequency distribution (or Histogram) of attribute knowledge. [2] The chart is known as once Italian economist/sociologist Vilfredo Pareto (1848-1923). It consists “a series of bars whose heights replicate the frequency or impact of problems. The bars area unit organized in descendent order of height from left to right. It states that regarding eighty percent of the issues come back from twenty percent of the causes and it’s extraordinarily helpful to spot the factors that have the best additive result on the system, and can able to classify them according the

weight of result to specialize in them.

3.4 Histogram

A Histogram is one of the most widely used and the most basic of the quality tools. It graphically summarizes large groups of data in an organized manner by clustering them all together in form of basic frequency distributions. A histogram mainly determines the shape of a data set.

3.5 Cause and Effect Diagram

The cause and effect diagram are also known as the Fishbone or the Ishiwaka diagram and its main purpose is to easily shows the defects and causes in a process. The main defect is listed in the arrow and the arrow branches out into the sub branches and detailed lists out different types of causes for such incidents. Fig 4 lists out the reasons for the margin of oversize due to which there are most rejections in a tape shank drill.

3.6 Control Chart

A control chart is one of the most important organizational tools in a standard quality review. The main purpose of it is to check whether the process is in control limits or not. Control chart typically works with two controls limits- Upper Control Limit (UCL) and Lower Control Limit (LCL). Control chart is the most important to tell about the process stability and variability. [1]

4. Data Collection

The following data was collected from a colleague who is working in a small-scale manufacturing company which produces different kinds of drill bits and material grinding. In Table 2, we have a check sheet data of rejected titanium drill bits and various reasons for the rejections.

Table 2. Check sheet for various modes of defects of Titanium HSS Drill bits while during Production

Table 3, as we can see shows the data of Lip Height variation in Titanium drill bits which are caused due to different MOS readings.

Table 3. Lip Height Reading

Data corresponding to the differences in MOS in mm measurement is show below in Table 4. We have 26 random readings for this purpose. This data will help configure the highest effect of MOS rejects.

Table 4. MOS Data

5. Statistical Analysis

5.1 Pareto Diagram

A pareto Chart is solely a frequency distribution (or Histogram) of attribute knowledge. [2] The chart is known as once Italian economist/sociologist Vilfredo Pareto (1848-1923). It consists “a series of bars whose heights replicate the frequency or impact of problems. The bars area unit organized in descendent order of height from left to right. It states that regarding eighty percent of the issues come back from twenty percent of the causes and it’s extraordinarily helpful to spot the factors that have the best additive result on the system, and can able to classify them according the

weight of result to specialize in them.

In figure 3 we can see that a pareto diagram is created by sportifying the values from the check sheet. This gives us a thorough information of the main causes of defects during the titanium HSS drill production.

Figure 2. Pareto Diagram – Defects in Taper Shank Drills

From the pareto chart we can observe that the top two defects are because of Margin oversize (MOS) and Body Diameter Test (BDT).

5.2 Histogram

A Histogram is one of the most widely used and the most basic of the quality tools. It graphically summarizes large groups of data in an organized manner by clustering them all together in form of basic frequency distributions. A histogram mainly determines the shape of a data set.

Table 5. Web-Centrality for MOS Data

Figure 3. Histogram

5.3 Cause and Effect Diagram

The cause and effect diagram are also known as the Fishbone or the Ishiwaka diagram and its main purpose is to easily shows the defects and causes in a process. The main defect is listed in the arrow and the arrow branches out into the sub branches and detailed lists out different types of causes for such incidents. Fig 4 lists out the reasons for the margin of oversize due to which there are most rejections in a tape shank drill.

Figure 4. Cause and Effect Diagram

5.4 Control Chart

A control chart is one of the most important organizational tools in a standard quality review. The main purpose of it is to check whether the process is in control limits or not. Control chart typically works with two controls limits- Upper Control Limit (UCL) and Lower Control Limit (LCL). Control chart is the most important to tell about the processstability and variability. [1]

Table 6. Control Chart Data Calculation

Figure 5. X-R Bar Control Chart

The above control chart indicates that the process is in control.

5.5 Run Chart

A Run Chart consist multiple data points and these data points shows process performance over time. This data points plotted in chronological order and sequence of graph shows that how process occurred.

Following are some advantages from using a Run Chart:

1. It monitors the process over time.

2. By using Run-Chart it helps to managers to take important decision when various problems are occurred.

Figure 6. Run-Chart for web Centrality

From the run chart we can see that even though the process is the control, it is very unstable. The process goes through a lot of variations.

5.6 Process Capability

As we know, a process is a unique combination of tools, materials, methods, and people which come together to produce a measureable output; for example a manufacturing line for machine parts.

The process capability is defined as “a measurable property of a process to the specification, expressed as a process capability index (e.g., Cpk or Cpm) or as a process performance index (e.g., Ppk or Ppm). The output of this measurement is usually illustrated by a histogram and calculations that predict how many parts will be produced out of specification (OOS).” (Pyzdek, T)

Figure 7. Process Capability Report

Immediately from the report we can see that the curve is centered which confirms the authenticity of the process. The overall and potential capability is closely aligned, this ensures that there will be no improvement if we work on shifts and drifts in the process. There are some nonconformities since the process spread is more than specification spread. Since Cpk is 0.43 which is less than 1.33, we can configure that the process needs an improvement.

6. Results and Discussion

After applying 7 Quality tools at Drill manufacturing industry we found the Root-Causes, quality level of drill, Process Performance, cost due to wastage and rejection of drills are mention below in Table.5 & Table.6.

Table 5. list of Root causes

Table 6. Result and Discussion

7. Conclusion and Proposed Solutions

This Paper describes how the quality tools were implemented in a manufacturing industry to analyze the reason of defects in a Titanium HSS Drill Bit production line. This Quality tools are used to monitor the overall Process of the production line and perform continuous process improvement. These basic quality tools were applied on Company data and analyzed to find the Root-causes. On the basis of root causes we are going to make action plan for further improvement. Process Capabilities studies was used using Minitab software to find the natural variation during production process. Process capability indices like Process capability ratio (Cp) and Process performance index (Cpk) will be useful for continuous process improvement.

As per the statistical analysis, Process Capabilities studies gave us an insight on where the process was wrong. The following improvements could be suggested based on that-

It would be appropriate to switch to Double margin drill bits instead of Single Margin drill bits. Since the diameter over size affects the single margin drill bits, it might not happen with double margin drill bits because of the minute calibrations which cannot go wrong.

We can notice from the run chart that the process has lots of extreme variations after a certain batch of runs. So it is necessary to calibrate the machine after a batch of atleast 20 parts. It will also help if the employee does a regular quick check up of the machine before part loading.

Tool Wear was another issue raised which affected the specs of the drill bits as even a tiny bit of wobble can cause lots of variations in a drill bit. For this purpose tool changing is suggested after every few batches as the machining causes lots of heat.

Human errors should always be avoided. Operators should be educated with instructions and manuals of such changes along with prescribed actions to work on all the changes.

Assigning more quality floor inspectors to check each batch of drill bits after the production. The defective pieces should be correctly tagged according to the caused defect.

Other factors like machine feed rate, cutting speed, torque and tool flank wear can be assigned to the R&D section for improvement after detailed analysis and reviews.

The Winslomatic machine is an older machine which is used in the company for Titanium coated and normal HSS drill bits. The rest of the newer drill bits are produced using the new advanced ANCA TX7 which has automatic pinpoint calibration is completely controlled by the software. Forgetting about the expenses and upgrading to ANCA TX 7 will be worth it for lowering the rate of rejections and scrap.

References

• Ciampa, Dan (1992). Total Quality: A User’s Guide for Implementation. Reading, Massachusetts: Addison-Wesley. p. xxii. ISBN 9780201549928. OCLC 634190702.
• Garvin, D. A., 1987, “Competing on the eight dimensions of quality”, Harvard Business Review, pp. 101-109
• Sebastião Nau, 2006. Acoustic noise in induction motors, IEEE Xplore, PCICON.2000.882782, pp. 1-29
• Donald S. Bruce, Cooper Tire & Rubber Company, Infinity QS – Case study of Cooper Tires. http://www.infinityqs.com/sites/infinityqs.com/files/files/Case%20Studies/InfinityQS-Case-Study-Cooper-Tire-and-Rubber-Company_Final.pdf
• Sourove Akther Momin, Khulna University of Engineering and Technology, Study on Quality Control Charts in a Pipe Manufacturing Industry, IJMEA-E20160806
• Freeman, J, G. Mintzas. 1999. Simulating c and u Control Schemes. The TQM Magazine. 11(4): 242–247.
• Young Thung Cho, 2017, Characterizing Sources of Small DC Motor Noise and Vibration. 2018 Feb 15;9(2). pii: E84. doi: 10.3390/mi9020084.
• Pyzdek, T, “Quality Engineering Handbook”, 2003, ISBN 0-8247-4614-7
• V. Parkash, D. Kumar, R. Rajoria, Statistical process control, International Journal of Research in Engineering and Technology 2 (8) (2013) 70-72.
• R. Srinivasu, G. Satyanarayana Reddy, S.R. Rikkula, Utility of quality control tools and statistical process control to improve the productivity and quality in an industry, International Journal of Reviews in Computing 5 (2011) 15-20.
• J.-H. Yang, M.-S. Yang, A control chart pattern recognition system using a satirical correlation coefficient method, Computer & Industrial Engineering 48 (2005) 205-221.
• Farzana Sultana, Nahid Islam and Razive Abdullahil Azeem, Implementation of Statistical Process Control (SPC) for Manufacturing Performance Improvement. Journal of Mechanical Engineering, Vol. ME 40, No.1, June 2009, Transaction of Mech Eng. Div., The Institution of Engineers, Bangladesh.