4 discussions due in 24 hours

combs

Chapter11QualityMethods.ppt

Home >Education homework help >4 discussions due in 24 hours

Quality
Improvement Methods

Chapter 11

11 | *

Learning Objectives

Describe three sources for quality improvement ideas: customer feedback, benchmarking, and employee feedback.

Describe qualitative quality improvement tools, including brainstorming, affinity diagrams, interrelationship diagrams, tree diagrams, process decision program charts, flowcharts, cause-and-effect diagrams, failure modes and effects analysis, and mistake-proofing.

Describe quantitative quality improvement tools, including inspection and sampling, check sheets, Pareto analysis, histograms, scatter diagrams, process capability analysis, run charts, and statistical process control charts.

11 | *

Two Scoops of Raisins in
a Box of Kellogg’s Raisin Bran

Cereal production is a highly automated process.
Kellogg’s uses quality management techniques such as establishing conformance standards, sampling, and statistical process control.
Statistical quality control charts are used to determine whether the variations observed from one cereal box to the next are random or have a specific cause.
Quality insurance inspectors periodically open random samples of the packed boxes that are ready to be shipped.

11 | *

Quality Improvement

When embarking on quality

improvement, the

Organization must:

Identify the goals and objectives of the program.

Identify the needs and preferences of the customers, users, or recipients of the processes that produce products and services.

Understand the current state of affairs at the organization.

Source: © Image Source/Corbis

11 | *

Sources of Quality Improvement Ideas

Customer Feedback
Benchmarking
Employee Feedback

11 | *

Customer Feedback

Customer feedback can come in many different formats:

Customer satisfaction

Customer choice analysis

Customer panels

Observation of recruited product users

Customer complaints

Unstructured information sources

11 | *

Customer Feedback Formats

Customer satisfaction: measurement of how well the products and services that a firm supplies meet or exceed its customers’ expectations
Customer choice analysis: an approach for quantifying the relative importance that customers assign to the various features and components of products and services
Customer panels: groups of individuals, selected according to predetermined criteria, who agree to provide periodic assessments of the quality of firms’ products and services
Blog: a web site where an individual writes a commentary about a specific product, issue, or topic

11 | *

Benchmarking

Benchmarking: a structured process for comparing the business practices of an organization to the best practices that can be identified in other organizations (partner firms, competitors, or suppliers) or other divisions within a company

11 | *

Table 11.1: Benchmarking Steps

11 | *

American Express Shares Benchmarked Quality Reports with Its Travel Service Clients

American Express operates one of the world’s largest travel agencies with $15.5 billion in worldwide travel sales in 2002.
AMEX has created a sophisticated back-office process that fully automates the quality control and data-gathering processes that companies need in order to manage travel.
The reports streamline and speed the process, save money for both AMEX and its clients, and enhance the quality of the corporate travel services.

11 | *

Employee Feedback

Employees in any organization are an excellent source of ideas for quality and process improvement.
Quality circle: a small volunteer group of employees responsible for similar or related work functions that meets regularly to identify, analyze, and solve quality and production problems related to its work
Specific tools and techniques are used to prioritize, analyze, and address these issues systematically.
These tools and techniques can be classified into two categories:

Qualitative

Quantitative

Source: © Image Source/Corbis

11 | *

Qualitative Quality Improvement Tools

Brainstorming
Affinity Diagram
Interrelationship Diagram
Tree Diagram
Process Decision Program Chart
Flowchart
Cause-and-Effect, Fishbone, or Ishikawa Diagram
Failure Modes and Effects Analysis (FMEA)
Mistake-proofing, Fail-safing, or Poka-Yoke

11 | *

Brainstorming

Brainstorming: a systematic method for generating a large number of creative problem-solving ideas in a relatively short amount of time based on input from many different individuals
Typically used in situations where a company needs to consider a broad range of solutions or options before coming to a conclusion
An excellent approach for arriving at a solution based on the collective wisdom of a group of individuals

11 | *

Affinity Diagram

Affinity diagram: a visual tool for organizing generated ideas into natural groups based on the collective wisdom of the participants.
To develop an affinity diagram:

Each viable idea generated during the brainstorming session is listed on a separate sticky note or card.

The notes are spread out so they can all be viewed.

The entire team gathers around the notes and moves them into logical subgroups until a consensus is reached.

11 | *

Interrelationship Diagram

Interrelationship diagram: a diagram that shows the connections and natural relationships between different ideas or constructs identified for quality improvement
Often used along with brainstorming and an affinity diagram

11 | *

Figure 11.3: Interrelationship Diagram

11 | *

Tree Diagram

Tree Diagram: a diagram used to describe how one idea branches into two or more subideas, each of which branches into further subideas and so on
Other names: systematic diagram, a tree analysis, an analytical tree, or a hierarchy diagram
Use:

Helps organizations shift analysis from general to specific ideas

11 | *

Figure 11.4: Tree Diagram

11 | *

Process Decision
Program Chart (PDPC)

Process decision program chart: a chart used to systematically identify initiatives that might potentially go wrong in a quality improvement plan that is under consideration
Once the potential problems are identified, countermeasures to prevent or offset the problems can be developed.
Particularly useful when the cost of failure is high

11 | *

Figure 11.5: Process
Decision Program Chart

11 | *

Cause-and-Effect,
Fishbone, or Ishikawa Diagram

Cause-and-effect diagram: a diagram used to illustrate the potential reasons for an outcome (usually a quality problem)
Use:

To structure and explore potential reasons for any undesirable outcome

11 | *

Figure 11.6: Cause-and-Effect Diagram

11 | *

Failure Modes and
Effects Analysis (FMEA)

Failure modes and effects analysis: a systematic approach for identifying all possible failures in a design, a manufacturing process, or a service process
A failure is any type of error or defect, especially one that affects the customer, and it can be either potential or actual.
The purpose of FMEA is to take action to eliminate or reduce failures, starting with the highest-priority ones.

11 | *

Mistake-Proofing,
Fail-Safing, or Poka-Yoke

Mistake-proofing: the use of any automatic device or method that either makes it impossible for an error to occur or makes the error immediately obvious once it has occurred
Use:

When process steps have been identified but human error can cause expensive or dangerous mistakes

11 | *

Recovering from Quality Mistakes: Sony Batteries in Dell Computers

Sony’s and Dell’s electronic products are widely used around the world.
In 2006, a handful of Dell customers reported that the batteries were overheating, catching fire, and even exploding.
Dell announced that it would replace 4.1 million lithium-ion batteries made by Sony.
The net cost of this recall was expected to be around $400 million.

11 | *

Quantitative
Quality Improvement Tools

Data-driven approaches commonly used in quality and process improvement efforts:

Inspection and Sampling

Check Sheet

Pareto Analysis and Bar Charts

Histogram

Scatter Diagram

Process Capability Analysis

Run Chart

Statistical Process Control Charts

11 | *

Inspection and Sampling

Companies use a combination of sampling and inspection techniques to collect and analyze data from a few selected outputs and attempt to generalize to the entire production batch.
Sampling: the statistical practice of selecting a few observations intended to generate inferences about a population of interest

11 | *

Sampling Plan Steps

Population of interest: a well-defined population of concern
Biased sample: won’t be representative of the entire group
Simple random sampling: each member of the population is equally likely to be selected as a sample
Stratified sampling: when the population of interest is believed to be made up of many subgroups or strata
Quota sampling: used to collect a prespecified number of observations from each subgroup
Sample size: making informed generalizations about the population of interest

11 | *

Type I Error

Type I error, alpha error, or false positive: the mistake of rejecting a hypothesis that should have been accepted

11 | *

Figure 11.7: Type I Error

11 | *

Type II Error

Type II error, beta error, or false negative: the mistake of failing to reject the hypothesis that everything within a population meets established standards when in reality the population does not conform to specifications

11 | *

Figure 11.8: Type II Error

11 | *

Check Sheet

Check sheet: a structured, prepared form for collecting and analyzing sampled observations

11 | *

Figure 11.9: Check Sheet

11 | *

Pareto Analysis and Bar Charts

Pareto analysis: a statistical technique for identifying and categorizing data based on frequency and percentage of occurrences

11 | *

Figure 11.10: Pareto Chart

11 | *

Histogram

Histogram: a type of graph used to show how often different values in a set of data occur

11 | *

Figure 11.11: Histogram

11 | *

Scatter Diagram

Scatter diagram: a graph of pairs of numerical data, with one variable on each axis, used to look for a relationship between the variables and identify potential correlations between them

11 | *

Figure 11.12: Scatter Diagram

11 | *

Process Capability Analysis

Process capability analysis: a method used to compare the statistical variations present within a system with any predefined existing range
Capable process: a process for which a prespecified percentage of measurements fall inside the specification limits

11 | *

When the Cp(k) = 1.0 the process variation is centered within the upper and lower tolerance limits. Process is capable of producing within specifications.

-1

-2

-3

Process Capability
Analysis/Capable Process

11 | *

When the Cp(k) > 1.0 the process variation

will be tighter with fewer defects

-1

-2

-3

Process Capability
Analysis/Capable Process

11 | *

Cp = Upper Specification – Lower Specification

6 σ

σ = std. deviation of the process mean

Calculating the Cp Ratio

11 | *

upper tolerance limit – X , X - Lower tolerance

Cpk = 3σ 3σ

Where X = 11

σ =.5

Cpk = min 11.7 – 11 , 11 – 10.3 = both = .47

(3)(.5) (3)(.5)

min

Calculating the Cpk Index

11 | *

Run Chart

Run chart: a simple graph used to display changes in observed data over a period of time

11 | *

Figure 11.13: Run Chart

11 | *

Statistical Process Control Charts

Statistical process control chart: a specific type of run chart calculated with well defined statistical properties

11 | *

Natural and Special-Cause Variations

Natural (or random) variations: variations observed in a process that appear to be random
Special-cause or assignable variation: a variation observed in a process that has a specific reason
Upper control limit: a line drawn at a prespecified distance (based on standard deviation units) above the mean value; the line represents the division between in-control and out-of-control processes
Lower control limit: a line similar to the upper control limit, but drawn at a prespecified distance below the mean value

11 | *

Figure 11.14: Statistical
Process Control Chart

11 | *

The Central Limit Theorem

Central limit theorem: a theorem stating that the shape of the distribution of averages for samples follows a normal distribution even if the original distribution was very different
Theoretical foundation for developing control charts

11 | *

Figure 11.15: Central Limit Theorem

11 | *

Control Chart for Variables

Variable data: process data measured on a continuous scale
x chart: a chart that shows how the process average changes over time
R chart: a chart that shows how the range of variations within a process change over time

11 | *

Control Chart for Attributes

Attribute data: the type of process data that cannot be measured on a continuous scale; the data are instead counted
p chart: a chart that tracks the changes in percentages of a specified attribute over a period of time

11 | *

Constructing SPC Charts

The construction of a statistical process control chart begins with the identification of a variable or attribute of interest and the selection of appropriate control chart.
Next, the appropriate sample size is selected to balance the probability of Type I and Type II errors with respect to the cost and effort associated with collecting the sample data.
The sampling time frame (days, hours, minutes, or some other measure) is decided.
The UCL and LCL are calculated, and the data are plotted.

11 | *

x Charts – Upper
and Lower Control Limits

UCL = X + A2 R

LCL = X – A2 R

Example:

UCL = 10 + .729 (.96) = 10 + .7 = 10.7

LCL = 10 - .729 (.96) = 10 - .7 = 9.3

(n = 4)

R = (1.4+1.5+.5+.7+1.3+.8+.5) / 8 = .96

11 | *

Figure 11.16: x Chart

11 | *

R Chart or Range Chart

Like an x chart, an R Chart requires calculating the data to be plotted, including the overall mean of the data plotted, the UCL, and the LCL.

11 | *

Table 11.4: Parameters for
Computing Control Limits for R Charts

Source: Quality Control of Materials (Philadelphia: American Society for Testing Materials, 1951).

11 | *

R Charts

UCLR = D4R

LCLR = D3R

Where: D4 = value from table

D3 = value from table

R = average of the ranges

Sample Size = 3

Example:

UCLR = (2.574)(.96) = 2.471

LCLR = (0)(.96) = 0

11 | *

Figure 11.17: R Chart

11 | *

Figure 11.18: p Chart

11 | *

Out-of-Control Process

Out-of-control process: a process in which one or more points in an SPC chart fall outside the control limits; out-of-control patterns can also be identified when process data exhibit certain specific trends even when all points fall within the control limits

11 | *

Table 11.6: Some
Commonly Used Out-of-Control Rules