UNIT-III STATISTICAL PROCESS CONTROL (SPC)

The seven tools of quality, Statistical Fundamentals – Measures of central Tendency and Dispersion, Population and Sample, Normal Curve, Control Charts for variables and attributes, Process capability, Concept of six sigma, New seven Management tools.

Introduction:

Most of the organizations use quality tools for various purposes related to controlling and assuring quality.

Although there are a good number of quality tools specific to certain domains, fields and practices, some of the quality tools can be used across such domains. These quality tools are quite generic and can be applied to any condition.

There are seven basic quality tools used in organizations. These tools can provide much information about problems in the organization assisting to derive solutions for the same.

A number of these quality tools come with a price tag. A brief training, mostly a self-training, is sufficient for someone to start using the tools.

Let's have a look at the seven basic quality tools in brief.

1. Flow Charts

This is one of the basic quality tool that can be used for analyzing a sequence of events.

The tool maps out a sequence of events that take place sequentially or in parallel. The flow chart can be used to understand a complex process in order to find the relationships and dependencies between events.

You can also get a brief idea about the critical path of the process and the events involved in the critical path.

Flow charts can be used for any field and to illustrate events involving processes of any complexity. There are specific software tools developed for drawing flow charts, such as MS Visio.

You will be able to freely download some of the open source flow chart tools developed by the open source community.

2. Histogram

Histogram is used for illustrating the frequency and the extent in the context of two variables.

Histogram is a chart with columns. This represents the distribution by mean. If the histogram is normal, the graph takes the shape of a bell curve.

If it is not normal, it may take different shapes based on the condition of the distribution. Histogram can be used to measure something against another thing. Always, it should be two variables.

Consider the following example: The following histogram shows morning attendance of a class. The X-axis is the number of students and the Y-axis the time of the day.

3. Cause and Effect Diagram

Cause and effect diagrams (Ishikawa Diagram) are used for understanding organizational or business problem causes.

Organizations face problems everyday and it is required to understand the causes of these problems in order to solve them effectively. Cause and effect diagrams exercise is usually a teamwork.

A brainstorming session is required in order to come up with an effective cause and effect diagram.

All the main components of a problem area are listed and possible causes from each area is listed.

Then, most likely causes of the problems are identified to carry out further analysis.

4. Check Sheet

A check sheet can be introduced as the most basic tool for quality.

A check sheet is basically used for gathering and organizing data.

When this is done with the help of software packages such as Microsoft Excel, you can derive further analysis graphs and automate through macros available.

Therefore, it is always a good idea to use a software check sheet for information gathering and organizing needs.

One can always use a paper-based check sheet when the information gathered is only used for backup or storing purposes other than further processing.

5. Scatter Diagram

When it comes to the values of two variables, scatter diagrams are the best way to present. Scatter diagrams present the relationship between two variables and illustrate the results on a Cartesian plane.

Then, further analysis, such as trend analysis can be performed on the values.

In these diagrams, one variable denotes one axis and another variable denotes the other axis.

6. Control Charts

Control chart is the best tool for monitoring the performance of a process. These types of charts can be used for monitoring any processes related to function of the organization.

These charts allow you to identify the following conditions related to the process that has been monitored.

·  Stability of the process

·  Predictability of the process

·  Identification of common cause of variation

·  Special conditions where the monitoring party needs to react

7. Pareto Charts

Pareto charts are used for identifying a set of priorities. You can chart any number of issues/variables related to a specific concern and record the number of occurrences.

This way you can figure out the parameters that have the highest impact on the specific concern.

This helps you to work on the propriety issues in order to get the condition under control.

Conclusion

Above seven basic quality tools help you to address different concerns in an organization.

Therefore, use of such tools should be a basic practice in the organization in order to enhance the efficiency.

Trainings on these tools should be included in the organizational orientation program, so all the staff members get to learn these basic tools.

Six Sigma

Introduction:- Six sigma methodology provides the techniques & tools to improve the capability & reduce the defects in any process. It was started by Motorola Company in 1987.

Six sigma means controlling defects to a very low level, so that the waste is the least and profits are more.

Definition: A business process that allows organizations to drastically improved their bottom line by designing and monitoring very day business activities in ways that minimize waste and resources while increasing customer satisfaction.

Need of Six Sigma:- An Aircraft consists of 10,000 different parts at 30 quality, 27 of those parts in an assembled aircraft would be defective. So, 30 quality level cannot be accepted as good enough so sigma level have to be increased by reducing the number of defectives. Even 40 quality level also is not sufficient for the aircraft case so to is preferred.

Concept of Six Sigma:-

i).Six sigma is similar to zero defect which is a standard of excellence. Proposed by Philip Crosby and zero defect is the objective of 60 principle.

ii).Six sigma strives for perfection. It allows 3.4 defects per million opportunities (99.999666% of accuracy). Defect can be anything from a faulty parts to an incorrect customer billing.

iii).Six sigma improves the process performance & decrease the variation & maintains consistent quality of the process output. This leads to defect reduction & improvements in profits, product quality & customer satisfaction.

Process of Six Sigma:-

There are five phases in the process of Six Sigma they are as follows;

1. Define:-

v  Determine Bench marks

v  Determine customer requirements

v  Get customer commitment

2. Measure:-

v  Develop defect measurement

v  Develop data collection process

v  Collect data

v  Compile the collected data

3. Analyse:-

v  Verify the data

v  Draw conclusion from data

v  Determine improvement opportunities

v  Determine root causes

4. Improve:-

v  Create improvement ideas

v  Set goals

v  Create problem statement

v  Create solution statement

v  Implement improvement methods

5. Control:-

v  Monitor improvement progress

v  Measure improvement statistically

v  Assess effectiveness

v  Make needed adjustments

6 3  3 6

¬ Specification Range ®

Motorola Prescribes 6 steps to achieve the six sigma:-

1. Identify the process you create or service you provide

2. Identify the customers for your product or service and determine what they consider

Important.

3.  Identify your needs to provide the product that satisfies the customer.

4.  Define the process for doing the work.

5.  Mistake – proof the process & eliminate waste effort

6.  Ensure continuous improvement by measuring, analyzing and controlling the improved process.

Scope of six sigma principle:-

Though the six sigma concept is originated from manufacturing field now it is applied to non-manufacturing processes also. Today six sigma can be applied to many fields such as services, medical & insurance procedures, call centres etc.,

Control Chart:-

It is a graph that displays data taken over time and the variations of this data. It is used to check whether the process is controlled statically or not. It is a graphical representation of measured actual process performance relative to computed control limits.

Objectives or Need of control charts:-

i)To provide the basis for taking decisions related to the acceptance or rejection of products being manufactured.

ii) To determine and eliminate the assignable causes of variations in a process.

iii) To constantly monitor a process to determine whether the process is controlled statically or not.

iv) To evaluate process stability and to decide when to adjust the process.

v) To improve the existing production procedures by analyzing the capability of the machine & the process .

vi) To establish & after the production procedure of any process, if necessary.

vii)To findout the natural capability of a production process, which permits better decisions on engineering tolerances.

viii)To make the possible diagnosis & correction of many production troubles & often brings substantial improvements in product quality and reduction of rework.

Types of control charts:-

I. Control chart for variables:-

A single measurable quality characteristic such as dimension, weight, volume, pressure, temperature, strength & hardness are called variables. Control chart for variables requires a measurement of the quality characteristic of the continuously varying magnitude. These variables can be measured and expressed in terms of units. Types of control chart for variables are;

i) Average (or) x Chart

ii) Range (or) R Chart

iii) Standard deviation (or) 0 chart

i)  X – Chart

v) MR (or) Moring Range Chart

II. Control chart for attributes:-

It is very difficult to represent many quality characteristics numerically. For example, colour, scratches, blow holes and damage in a component. The characteristics cannot be measured numerically but can be classified based on conforming items or non conforming items to the given specifications. Control chart for attributes require a determination of whether the inspected parts may be accepted if they are conforming to the specifications (or) rejected if they are non conforming to the specifications. Types of control chart for attributes are;

i)  P – Chart

ii)  np – Chart

iii)  C – Chart

iv)  U – Chart

Three Zones in Control Charts:

Zone I is called stable zone. It is up to 2 limits on both sides. If the performance is in this zone, then the production may be continued without any modifications.

Zone II is called warning zone. It is between 2 and 3. If the performance is continuously falling in this zone, then the process may require adjustment to bring the performance to Zone I.

Zone III is called action zone. If the performance lies in this zone, then it calls for immediate action to adjust the process.

¬ Variable or Attribute ® / -3
-2
-1

+1
+2
+3 / III

II

I

II

III
Time ®

Types of Control Charts for Variable Data:

Chart Name / Central Line / Control Limits / Remarks
- Chart / / Control limits are:

i.e., UCL =
LCL = /

R – Chart / / UCL =
LCL = / Where D3 & D4 are statistical factors (constants) obtained from table.
X – Chart / /
i.e., UCL =
LCL = / Where is the average of moving ranges.
 – Chart / / (a) If sample SD is known:
UCL =
LCL = / Where B1, B2, B3 & B4 are statistical factors (constants) obtained from table and
(b) If population SD 1 is known:
UCL =
LCL =
MR - Chart / / UCL = 3.276 ´
LCL = 0 /

Types of Control Charts for Attribute Data:

Chart Name / Central Line / Control Limits / Remarks
np – Chart / n / Control limits are:

i.e., UCL =
LCL = /
p = , Fraction Defective
np – No. of defectives
n – No. of items inspected in a subgroup
p – Chart / / Control limits are:

i.e., UCL =
LCL =
c – Chart / / Control limits are:

i.e., UCL =
LCL = / c – No. of defects
n – No. of samples
, Average no. of defects
u – Chart / / Control limits are:

i.e., UCL =
LCL = / u – No. of defectives per unit in a sample
=
i.e., =

Applications of control charts:-

v  It detects the variation in processing & warns if there is any deviation from the specified tolerance limits.

v  It is widely used in solving problems related to process stability & improvement of capability by reducing the variability in the process adopted.

v  It is used for controlling the characteristics of any kind of processes in order to improve the quality of the product.

v  It means possible the diagnosis and correction of the many production troubles.

v  The control chart is used to evaluate process stability & to decide when to adjust the process.

Construction of Control Chart:-

/ UCL
CL
LCL

The seven statistical tools of quality