MFE 3100 © Quality Management and Control E.Francalanza · MFE3100 Quality Management and Control...

MFE3100

Quality Management and Control

© E. Francalanza 1

© E.Francalanza MFE 3100 – Quality Management and Control


• Set your Mobile Phones to Silent

• Respect your colleagues – no private

conversations

• Be on time

MFE3100


© E. Francalanza 2


• SPC for Attribute Data

• Tutorial Overview


MFE3100


© E. Francalanza 3



• The quality of many products and services is dependent upon characteristics which cannot be measured as variables.

• These are called attributes and may be counted, having been judged simply as either present or absent, conforming or non-conforming, acceptable or defective

• Attribute Chart data is more easily assessed. Variables are sometimes converted to attributes for assessment.

• Attributes are not so sensitive a measure as variables, and therefore detection of small changes is less sensitive.

MFE3100


© E. Francalanza 4


Attribute Data

Non-Conforming

Units

Non-Conformities


• Non-Conforming Units

– Can be wholly described as failing or not

failing, acceptable or defective, present or not

present.

– Examples: Ball-Bearings, Invoices, Workers.

• For 100 Ball-Bearings we can state how many are

defective or non-conforming. If 2 Ball-Bearings are

classified as unacceptable or defective, 98 will be

acceptable.

MFE3100


© E. Francalanza 5


• Non-Conformities

– Used to describe a product or service.

– Examples: Errors, Sales Calls, Truck

Deliveries.

• For a product such as a Windscreen which is being

examined for defects such as scratches or

bubbles, one can only measure the number of non-

conformities present, and one cannot imply on

defects which are not present.


Name Sample Size Defect Type

The p-chart Varied Defective

The np-chart Constant Defective

The u-Chart Varied Defects

The c-Chart Constant Defects

MFE3100


© E. Francalanza 6


Check on

sample size (n)

Is n

constant

?

Check on type

of item

Is the item

a defective

unit rather

than a

defect?

Check on type

of item

Is the item

a defective

unit rather

than a

defect?

u-Chart

No Yes

p-Chart c-Chart np-Chart

No Yes No Yes


• If process control is going to be effective, it

is off utmost importance to clarify what

constitutes a defective, non-conformance,

defect or error.

• Judgments can vary, leading to heated

discussions, therefore control samples

have to be set up.

MFE3100


© E. Francalanza 7



• Most Popular Chart

• Used for process control of defective units,

when it is not possible to take a sample of

a constant sample size

• Data Required:

– Sample Size

– No. Of Defectives per Sample

MFE3100


© E. Francalanza 8



• Company ABC wants to carry out an SPC

study on the delivery of Textile

Components (ex. Shirts).

• The Textiles are delivered in varying

batches. Samples are taken in proportion

to the batches, and the number of rejects

(ex. due to tears) is recorded for 24

deliveries.

• Data is presented in the next slide.

MFE3100


© E. Francalanza 9


Issue Size Number of

Rejects Proportion

Defective 1135 10 0.009

1405 12 0.009

805 11 0.014

1240 16 0.013

1060 10 0.009

905 7 0.008

1345 22 0.016

980 10 0.010

1120 15 0.013

540 13 0.024

1130 16 0.014

990 9 0.009

1700 16 0.009

1275 14 0.011

1300 16 0.012

2360 12 0.005

1215 14 0.012

1250 5 0.004

1205 8 0.007

950 9 0.009

405 9 0.022

1080 6 0.006

1475 10 0.007

1060 10 0.009

p

Mean: Mean:


MFE3100


© E. Francalanza 10



• Similar to a p-chart but the sample size is

constant.

• The data values plotted are the actual

number of defective items per sample

rather than the proportion.

MFE3100




• A manufacturer of ball-point pens takes 50

samples of size 100 are taken every hour

from the production process to check for

defective products.

• The number of defects found in each

sample is recorded using a tally chart and

a Histogram is plotted with the results


MFE3100




Sample No. No of Defective

Parts per Sample Sample No.

No of Defective

Parts per Sample

1 2 26 0

2 4 27 3

3 1 28 1

4 0 29 2

5 0 30 1

6 4 31 2

7 5 32 1

8 3 33 5

9 2 34 3

10 3 35 0

11 2 36 2

12 3 37 2

13 0 38 1

14 3 39 3

15 1 40 1

16 2 41 1

17 3 42 3

18 1 43 0

19 2 44 2

20 1 45 1

21 2 46 2

22 4 47 0

23 2 48 4

24 1 49 2

25 6 50 1

Total Defective:

Step 1: Calculate

Total No. of

Defectives 100


p = 0.02

n = 100.00

np = 2.00

UCL 6.40

LCL 0.00

Step 2: Calculate np

Step 3: Calculate

UCL & LCL

MFE3100




0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

1 2 3 4 5 6 7 8 9 1011121314151617181920212223242526272829303132333435363738394041424344454647484950

Step 4: Plot Graph


MFE3100




• The u-chart is used for process control for defects when it is not possible to take a sample of constant sample size. This is similar to the p-chart, where the data values plotted on the chart are the proportion of faults per sample.

• Note however the difference that it is the number of non-conformities per item within the sample that is being monitored, and not the number of items per sample which have been rejected.


MFE3100




• Considering a process which manufactures cooling fan blades.

• Each blade has a possible 17 measurements made on it. If variable charts had to be plotted and analyzed in detail this would take up a lot of time.

• Instead dimensions read are compared to the tolerance specifications and the number of defects (i.e. dimensions out of tolerance) are listed.

• The next table shows data gathered from 20 different batches.


Number of Blades in

Sample Number of defects in

Sample Number of Defects

per Unit (u)

85 40 0.47

88 82 0.93

92 95 1.03

83 78 0.94

78 125 1.60

75 50 0.67

80 105 1.31

72 35 0.49

80 72 0.90

92 85 0.92

75 68 0.91

81 77 0.95

43 75 1.74

80 46 0.58

125 120 0.96

120 105 0.88

155 250 1.61

81 152 1.88

45 17 0.38

50 13 0.26

u

Mean: Mean:

𝒖

MFE3100





MFE3100




• The c-chart is used for process control of

defects when it is possible to take samples

at a constant sample size.

• The data which is plotted on the chart are

the number of defects c in each sample.


• In a polythene film process, the number of

defects – fisheyes – on each identical

length of film are being counted.

• The following chart portrays the variation

of the number of fisheyes which have

been found on inspecting 50 lengths,

randomly selected, over a 24-hour period.

MFE3100





• If the chart is now used to control the

process, we may examine what happens

over the next 25 lengths, take over a

period of 12 hours.

MFE3100





MFE3100




• What elements make up the main

components of the ISO 9001 Management

System, and how does this system work?


• A laptop manufacturer sells 10,000 products per year. The costs for

the products sold are detailed in the table below. The products are

sold at the price of €1,200 per laptop. Analyze these costs. All the

costs listed are in (x103) €.

MFE3100





Preventive € 500,000

Appraisal € 700,000

Failure € 700,000

Cost of Quality € 1,900,000 24%

Manufactuirng € 6,050,000 76%

Total Cost € 7,950,000

No. Sales 10000

Price/unit € 1,200

Sales € 12,000,000

Cost € 7,950,000

Profit € 4,050,000 34%

MFE3100




• What are the types of variation that are encountered when analyzing a process?

– Common causes of variation

– Special causes of variation

• Give some examples of common and special causes of variation.

– Common causes – inherent process instability, such as vibrations.

– Special causes – chattering, tool wear, mis-alignments


• Why do we need to use these ‘old’ quality tools in problem solving?

– Promotes a Systematic approach

– Aids in visualization of problems

– Facilitate data gathering

• Why is it important to identify the root-cause of a problem?

– If the root cause of a problem is not addressed, there is a very high probability that the problem will re-surface.

MFE3100




• Why do we need to use these ‘new’ quality tools in problem solving? – Induces teams to think about alternative causes

of problems.

– Explores possible relationships between complex problems and their causes.

– Aid planning and implementation of solutions.

• Which tools would you use to plan and implement solutions? – Process Decision Programme Chart (PDPC)

– Arrow Diagram


• What is the main Kaizen philosophy?

– Small continuous improvements which lead to Large

overall improvement of the company.

• Mention some main differences between Kaizen

and Innovation.

MFE3100




• Use the following data in the table to plot an SPC chart by variable. The specification for the dimension is given as 29.5±0.5

– Is the Process in Statistical Control?

– Is the Process Capable?


Grand or Process Mean

Mean Range Step 1: Establish Process Mean and Range

MFE3100




Mean Chart:

A2 1.023

UCL 29.92

LCL 29.50

Range Chart:

D3 0

D4 2.574

LCL 0

UCL 0.532

Step 2:

Establish Control Limits


UCL

LCL

Step 3:

Plot Variable Chart

MFE3100




UCL

LCL

Step 4:

Plot Atribute Chart


USL 30.00

LSL 29.00

Cp 1.363

𝜎 =𝑅

𝑑2

Step 5:

Check for

Process Capability Calculate σ

Calculate Cp

R 0.21

d2 1.693

σ 0.122

MFE3100




MFE 3100 © Quality Management and Control E.Francalanza · MFE3100 Quality Management and Control...

Documents

Transcript of MFE 3100 © Quality Management and Control E.Francalanza · MFE3100 Quality Management and Control...