21
How to use Statistical Process Control (SPC) charts?

How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

  • Upload
    others

  • View
    29

  • Download
    1

Embed Size (px)

Citation preview

Page 1: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

How to use Statistical Process Control (SPC) charts?

Page 2: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

ContentsWhy do we use statistical process control (SPC) charts? PAGE 1

What is a statistical process control (SPC) chart? PAGE 2

How to choose between a line chart, run chart and control chart? PAGE 4

The different type of variation that exists within a process/system PAGE 6

What to do when seeing non-random/special cause variation PAGE 7

When do you revise the centre line and control limits? PAGE 8

Formatting recommendations when creating an SPC PAGE 9

APPENDIX 1 – Control (Shewhart) chart selection guide PAGE 10

APPENDIX 2 – Flowchart for selecting the most appropriate control (Shewhart) chart PAGE 11

APPENDIX 3 – Rules of Non-Random Variation for a Run Chart PAGE 14

APPENDIX 4 – Rules of Special Cause for a Control Chart PAGE 16

APPENDIX 5 – Chart examples PAGE 17

Page 3: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

Why do we use statistical process control (SPC) charts?

Statistical Process Control (SPC) charts are used to study how a system or process changes over time. It allows us to understand what is ‘different’ and what is the ‘norm’. By using these charts, we can then understand where the focus of work needs to be concentrated in order to make a difference.

We can also use SPC charts to determine if an improvement is actually improving a process and also use them to ‘predict’ statistically whether a process is ‘capable’ of meeting a target.

SPC charts are therefore the best tools to determine:

• The variation that lives in the process.• Predict how the process will perform in the future.• If our improvement strategies have had the desired effect. • Determine if an improvement strategy has sustained the gains.

Page 1Contents PageProduced by Forid Alom

Page 4: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

What is a statistical process control (SPC) chart?

Statistical Process Control (SPC) charts consist of data over time and come in two forms:1. Run charts2. Control chart (also known as a Shewhart chart).

Figure 1a and 1b illustrate the elements of both charts.

Figure 1a – Elements of a run chart. Data plotted over time with a centre line based on the median value

Figure 1b – Elements of a control chart. Data plotted over time with a centre line based on the mean value, an upper control limit and a lower control limit.

Me

asu

re

Time Time

Me

asu

rePage 2Contents Page

CL = Centreline(Mean)

Produced by Forid Alom

Page 5: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

What is a statistical process control (SPC) chart?

Statistical Process Control (SPC) charts consist of data over time and come in two forms:1. Run charts – can take any data type i.e. count, percentage, rate, days between etc

2. Control chart (also known as a Shewhart chart) – each data type has a specific chart

There is only one type of run chart and many different types of control charts (figure 2).

SPC CHARTS

RUN CHART CONTROL CHARTS

C Chart

U Chart

U ‘ Chart

P Chart

P ‘ Chart

G Chart

T Chart

XmRChart

X bar S Chart

Figure 2 – Different type of SPC charts

Page 3Contents PageProduced by Forid Alom

Page 6: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

How to choose between a line chart, run chart and control chart?

When constructing an SPC chart, you need to make sure you have the correct number of data points. Figure 3 explains how to choose the right chart?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24+

Line Chart Run Chart Control chart with “trial” limits

Control chart

"Trial" UCL

CL = Centreline

"Trial" LCL

UCL

CL = Centreline

LCL

CL = Centreline

Figure 3 – Choosing the right chart depending on the number of data points. When you have less than 20 data points, you can create a control chart with “trial” limits. Once you reach 20

points, you would recalculate the limits again.

Number of data points

Page 4Contents PageProduced by Forid Alom

Page 7: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

How to choose between a line chart, run chart and control chart?

When using a control chart, it is important you choose the right one. There are different types of control charts depending on what type of data you’re looking at. Appendix 1 and Appendix 2 provide guidance on choosing the correct chart.

Control chart with “trial” limits

Control chart

"Trial" UCL

CL = Centreline

"Trial" LCL

UCL

CL = Centreline

LCL

C Chart

U Chart

U ‘ Chart

P Chart

P ‘ Chart

G Chart

T Chart

XmRChart

X bar S Chart

Page 5Contents PageProduced by Forid Alom

Page 8: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

The different type of variation that exists within a process/system

By ordering the data points over time, we can identify the variation that exists within process/system. Figure 4 describes how we describe these.

Random / Common cause variation Non-random / Special cause variation

Is inherent in the design of the process Is due to irregular or unnatural causes that are not inherent in the design of the process

Is due to regular, natural or ordinary causes Affect some, but not necessarily all aspects of the process

Affects all the outcomes of a process Results in an “unstable” process that is not predictable

Results in a “stable” process that is predictable Also known as non-random or assignable variation

Also known as random or unassignable variation

SPC CHARTS

RUN CHARTS CONTROL CHARTS

Random Variation

Non-Random Variation

Common Cause Variation

Special Cause Variation

Figure 4 – Different type of variation that exists within a process/system Page 6Contents PageProduced by Forid Alom

Page 9: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

What to do when seeing non-random/special cause variation

There are specific rules that help us identify non-random variation (Appendix 3) / special cause variation (Appendix 4) within an SPC charts.

When we see any of these rules within a chart, we need to take the following steps:

Investigate what caused it? Was it due to internal (i.e. change in process) or external factors (i.e. cyber attack)?

Determine if any action is needed

Revise centreline and control limits (if appropriate)

1

3

2

Page 7Contents PageProduced by Forid Alom

Page 10: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

When you change the operational definition.

When the control chart remains unstable for 20 or more subgroups and approaches to

identify and remove the special causes have been exhausted.

When do you revise the centre line and control limits?

Below are five reasons that warrant recalculating of limits.

When “trial” limits have been calculated with fewer than 20 subgroups.

When the initial control chart has special causes and there is a desire to use the

calculated limits for analysis of data to be collected in the future

When improvements have be made to the process and the improvements result in

special causes on the control chart.

1

5 2

34

Page 8Contents PageProduced by Forid Alom

Page 11: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

Formatting recommendations when creating an SPC

When creating SPC charts at ELFT, please note the following recommendations regarding formatting.

X-axis and Y-axis clearly

labelled

Title of chart explains what you’re measuring and the type of chart you’re using

Don’t use bright colours for data points. As a standard,

ELFT uses Blue

Don’t use bright colours for centreline. As a standard,

ELFT uses Black

Centreline clearly labelled and value added. Rounded to

3 decimal places

Limits clearly labelled

25.039%

25.039%

UCL

LCL

0%

10%

20%

30%

40%

50%

60%

05

-Fe

b-1

8

12

-Fe

b-1

8

19

-Fe

b-1

8

26

-Fe

b-1

8

05

-Mar

-18

12

-Mar

-18

19

-Mar

-18

26

-Mar

-18

02

-Ap

r-1

8

09

-Ap

r-1

8

16

-Ap

r-1

8

23

-Ap

r-1

8

30

-Ap

r-1

8

07

-May

-18

14

-May

-18

21

-May

-18

28

-May

-18

04

-Ju

n-1

8

11

-Ju

n-1

8

18

-Ju

n-1

8

25

-Ju

n-1

8

02

-Ju

l-1

8

09

-Ju

l-1

8

16

-Ju

l-1

8

23

-Ju

l-1

8

30

-Ju

l-1

8

06

-Au

g-1

8

13

-Au

g-1

8

20

-Au

g-1

8

27

-Au

g-1

8

Did

No

t A

tte

nd

/ %

Week

Number of people who did not attend their appointment - P Chart

Annotations added to chart to explain special cause

Limit recalculation based on the five reasons explains in

previous slide

PDSA 1

Page 9Contents PageProduced by Forid Alom

Page 12: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 1 – Control (Shewhart) chart selection guide

Type of data Continuous (variables data)

- Data in the form of a measurement- Requires some type of scale- Time, money, physical measure (i.e. length, height, weight, temperature) and throughput (volume of workload, productivity)

Count or Classification (Attributes data)

- Numbers of items that passed or failed- Data must be whole number when originally collected (can’t be fraction or scaled data)- This data is counted, not measured

Count- 1,2, 3, 4 etc. (errors, falls, incidents)- Numerator (incidences) can be greater than denominator (opportunities)

Classification- Either/or, pass/fail, yes/no- Percentage or proportion- Numerator cannot be greater than denominator

Equal area of opportunity

Unequal area of opportunityEqual or unequal

subgroup size

C chart P chart

Each dot on the chart consists of a single observation (i.e. cost per episode of care, waiting time for each patient)

Subgroup size of 1 (n=1)

Each dot on the chart consists of multiple data values

- The X-bar chart is the average of multiple data points (i.e., average cost per case for all cases this week, average waiting time for a specific no. of ER patients [e.g. 5])- S plots the standard deviation of the data values

Equal or unequal subgroup size (n>1)

XmR chart X and S chart

Subgroup size <5000

Subgroup size > 5000

Subgroup size < 5000

Subgroup size > 5000

P’ chartU chart U’ chart

Other types of control charts for count/classification data:• T Chart [time between rare events]• G Chart [case between rare events]

Other types of control charts for continuous data:• X-bar and Range Chart• Median and Range Chart Page 10Contents Page

Produced by Forid Alom

Page 13: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 2 – Flowchart for selecting the most appropriate control (Shewhart) chart

*A run chart may be used with any type of

data. It is often the starting point for viewing data over time when little data are yet available.

Do you have

Continuous data?

Do you have

Attribute data?

DATA*

A B

Get in touch with QI team

Yes

Yes

No

No

Attribute data can only take particular values. There may potentially be an infinite number of those values, but each is distinct and there’s no grey area in between. The data can be numeric – for example number of falls – but it can also be categorical – such as pass or fail, male or female, good or bad.

Important points about attribute data:

•It is counted, not measured.•Data must be whole numbers when collected (can’t be fraction or scaled data)•There is two sub-types:

-Count data-Classification data

Typical examples:

•Number of falls•Number of violent incidents•% of missed doses•% of service users scoring “effective” or “very effective” for patient care

Continuous data is not restricted to defined separate values, but can have almost any numeric value and can be subdivided into finer and finer increments, depending upon the precision of the measurement system. Between any two continuous data values, there may be an infinite number of others – for example time waited for 1st appointment.

Important points about continuous data:

•Data is in the form of a measurement. -Time-Money-Physical measure (length, height, weight, temperature)-Throughput (volume of workload, productivity)

•It requires some type of scale

Typical examples

•Waiting times for 1st appointment•Service user length of stay•Service user weight

No

Page 11Contents PageProduced by Forid Alom

Page 14: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 2 – Flowchart for selecting the most appropriate control (Shewhart) chart

Count

- 1, 2, 3, 4 etc. (errors, occurrences, defects,

complications)

- Numerator can be greater than denominator

Typical examples:

- Number of falls

- Number of incidents of physical violence

- Complaints per 1,000 visits

Classification

- Either/or, pass/fail, yes/ no

- Percentage or proportion

- Numerator cannot be greater than the

denominator

- Can have an equal or unequal subgroup size

Typical examples:

- Did Not Attend (DNA) rate

- % of service user participation

- % of safety huddles completed every

week

A Do you have

Count data?

YesDo you have

Classification

data?

No

Get in touch

with QI team

PChart

Yes

E.g. Number of Falls causing

harm as a % of all falls

reported

NoYes

TChart

E.g. Days between

number of Falls

Do you have an equal area of

opportunity?

Typical examples:

- Number of incidents per week

- Number of uses per month

NoYesCChart

UChart

E.g. Number of Falls per

1,000 occupied bed days

E.g. Number of Falls

Is it a rare

event?

Are you

measuring in

days?

Yes

No

GChart

E.g. Cases

between infection

No

Page 12Contents PageProduced by Forid Alom

Page 15: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 2 – Flowchart for selecting the most appropriate control (Shewhart) chart

B

Does each data point on the

chart consist of a single

observation?

Typical examples:

- Cost per episode of care

- Waiting time for each patient

- Average decibel reading for noise level

Typical examples:

- Average waiting time for 1st appointment

across multiple teams

- Average cost per case for all cases this week

- Average weight gain for all service users this

month

YesX Bar S

Chart

X Bar S chart characteristics:

- Two charts are created:

o An average chart known as the X bar chart

▪ Upper and lower control limit vary with sample size

▪ Y-axis usually the average of a measurement

o A standard deviation chart known as the S chart

▪ Y-axis is the standard deviation of all data points making up each point on the X bar chart

E.g. Cost per episode of Falls E.g. Average cost per episode of

falls across all inpatient wards

XmRChart

No

Page 13Contents PageProduced by Forid Alom

Page 16: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 3 – Rules of Non-Random Variation for a Run Chart

There are four rules for a run chart that help you identify non-random variation.

CL = MEDIAN

20

30

40

50

60

70

80

90 RULE 1 - SHIFT

CL = MEDIAN

15

20

25

30

35

40

45RULE 2 - TREND

CL = MEDIAN

10

15

20

25RULE 3 - TOO MANY OR TOO FEW

CL = MEDIAN

35

55

75

95

115RULE 4 - ASTRONOMICAL POINT

Rule 1 – Shift

Six or more consecutive points either all above or all below the centreline (CL). Values that fall on

the CL do not add to nor break a shift. Skip values that fall on the median and continue counting

Rule 2 – Trend

Five or more consecutive points all going up or all going down. If the value of two or more

successive points is the same (repeats), ignore the like points when counting.

Rule 3 – Too Many or Too FewIf there are too many or too few runs, this is a sign of non-random variation. To see what an appropriate number of runs for a given number of data sets, refer to the table on the next page. An easy way to count the number of runs is to count the number of times the line connecting all the data points crosses the median and add one. If the number of runs you have are:- Within the range outlined in the table, then you have a random pattern.- Outside the range outline in the table, then you have a non-random pattern or signal of change.

Rule 4 – Astronomical point

This is a data point that is clearly different from all others. This is a judgement call. Different

people looking at the same graph would be expected to recognise the same data point as

astronomical.

Data line crosses the centreline (CL) once.

10 data points = runs should be between 3 and 9

Total runs = 2 so too few runs

Page 14Contents PageProduced by Forid Alom

Page 17: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 3 – Rules of Non-Random Variation for a Run Chart

Page 15Contents PageProduced by Forid Alom

Table 1 - Runs Rule Guidance – Table for checking too many or too few runs on a Run chart

Total no. of data points on run chart

not falling on

Lower limit for no. of runs (< than

this is “too few”)

Upper limit for no. of runs

(>than this is “too many”)

Total no. of data points on run chart

not falling on

Lower limit for no. of runs (< than

this is “too few”)

Upper limit for no. of runs (>than

this is “too many”)

10 3 9 37 13 25

11 3 10 38 14 26

12 3 11 39 14 26

13 4 11 40 15 27

14 4 12 41 15 27

15 5 12 42 16 28

16 5 13 43 16 28

17 5 13 44 17 29

18 6 14 45 17 30

19 6 15 46 17 31

20 6 16 47 18 31

21 7 16 48 18 32

22 7 17 49 19 32

23 7 17 50 19 33

24 8 18 51 20 33

25 8 18 52 20 34

26 9 19 53 21 34

27 10 19 54 21 35

28 10 20 55 22 35

29 10 20 56 22 36

30 11 21 57 23 36

31 11 22 58 23 37

32 11 23 59 24 38

33 12 23 60 24 38

34 12 24

35 12 24

36 13 25

Page 18: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 4 – Rules of Special Cause for a Control Chart

There are five rules for a control chart that help you identify special cause variation.

UCL

CL = MEAN

LCL25

30

35

40

45

50

55

60

65

70 RULE 1 - SHIFT UCL

CL = MEAN

LCL21

26

31

36

41

46

51

56

61RULE 2 - TREND

UCL

CL = MEAN

LCL21

26

31

36

41

46

51

56

61

RULE 4 - TWO OUT OF THREE

UCL

CL = MEAN

LCL21

26

31

36

41

46

51

56

61

66 RULE 3 - THREE SIGMA VIOLATION

UCL

CL = MEAN

LCL21

26

31

36

41

46

51

56

61

RULE 5 - 15 OR MORE POINTS HUGGING THE MEAN

Rule 1 – Shift

Eight or more consecutive

points either all above or all

below the centreline (CL).

Values that fall on the CL do

not add to nor break a shift.

Skip values that fall on the

mean and continue counting

Rule 2 – Trend

Six or more consecutive

points all going up or all

going down. If the value

of two or more successive

points is the same

(repeats), ignore the like

points when counting.

Rule 3 – Three Sigma

Violation

When you have a data

point that exceeds the

UCL/LCL.

Rule 4 – Two out of

three

When you get two out

of three consecutive

points in the outer

one-third of chart.

Rule 5 – 15 or more data points hugging the mean

15 or more data points hugging the centreline (inner one-third of the chart).

In a normal distribution, you should have around 68% of the data near the

mean of the distribution (+/- 1 standard deviation). When you get a pattern

like this, you’re exceeding the 68%.

Page 16Contents PageProduced by Forid Alom

Page 19: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

APPENDIX 5 – Chart Example 1 (Count Data)

Page 17Contents Page

Date No. of Incidents

6-Jan-2014 8220-Jan-2014 693-Feb-2014 72

17-Feb-2014 533-Mar-2014 82

17-Mar-2014 9331-Mar-2014 7114-Apr-2014 6328-Apr-2014 5812-May-2014 4226-May-2014 68

9-Jun-2014 6323-Jun-2014 877-Jul-2014 62

21-Jul-2014 544-Aug-2014 65

18-Aug-2014 611-Sep-2014 73

15-Sep-2014 5829-Sep-2014 65

30

40

50

60

70

80

90

100

6-J

an-1

4

20-

Jan

-14

3-F

eb-1

4

17-

Feb

-14

3-M

ar-1

4

17-

Mar

-14

31-

Mar

-14

14-

Ap

r-14

28-

Ap

r-14

No

. of

Inci

den

ts /

Co

un

t

Number of incidents - Line chart

70.000

30

40

50

60

70

80

90

100

6-J

an-1

4

20-

Jan

-14

3-F

eb-1

4

17-

Feb

-14

3-M

ar-1

4

17-

Mar

-14

31-

Mar

-14

14-

Ap

r-14

28-

Ap

r-14

12-

May

-14

No

. of

Inci

den

ts /

Co

un

t

Number of incidents - Run Chart

67.933

UCL92.660

LCL

43.207

30

40

50

60

70

80

90

100

6-J

an-1

4

20-

Jan

-14

3-F

eb-1

4

17-

Feb

-14

3-M

ar-1

4

17-

Mar

-14

31-

Mar

-14

14-

Ap

r-14

28-

Ap

r-14

12-

May

-14

26-

May

-14

9-J

un

-14

23-

Jun

-14

7-J

ul-

14

21-

Jul-

14

No

. of

Inci

den

ts /

Co

un

t

Number of incidents - C Chart (Trial Limits)

67.050

UCL91.615

LCL42.485

30

40

50

60

70

80

90

100

6-J

an-1

4

20-

Jan

-14

3-F

eb-1

4

17-

Feb

-14

3-M

ar-1

4

17-

Mar

-14

31-

Mar

-14

14-

Ap

r-14

28-

Ap

r-14

12-

May

-14

26-

May

-14

9-J

un

-14

23-

Jun

-14

7-J

ul-

14

21-

Jul-

14

4-A

ug-

14

18-

Au

g-14

1-S

ep-1

4

15-

Sep

-14

29-

Sep

-14

No

. of

Inci

den

ts /

Co

un

t

Number of Incidents - C Chart

Type of data

Variables dataAttributes data

Multiple Observations

Single Observation

ClassificationCount

Here is an example of the SPC charts you’d expect when measuring count data i.e. No. of incidents.

Produced by Forid Alom

Page 20: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

30.701%

UCL

35.515%

LCL

25.886%

20%

22%

24%

26%

28%

30%

32%

34%

36%

38%

40%

Jan

-14

Feb

-14

Mar

-14

Ap

r-14

May

-14

Jun

-14

Jul-

14

Au

g-14

Sep

-14

Oct

-14

No

v-14

Dec

-14

Jan

-15

Feb

-15

Mar

-15

Ap

r-15

May

-15

Jun

-15

Jul-

15

Au

g-15

Sep

-15

Oct

-15

No

v-15

Dec

-15

Did

No

t A

tten

d /

%

% of Did Not Attend (DNA) Appointments - P Chart32.500%

UCL

37.103%

LCL

27.898%

20%

22%

24%

26%

28%

30%

32%

34%

36%

38%

40%Ja

n-1

4

Feb

-14

Mar

-14

Ap

r-14

May

-14

Jun

-14

Jul-

14

Au

g-14

Sep

-14

Oct

-14

No

v-14

Dec

-14

Jan

-15

Feb

-15

Mar

-15

Did

No

t A

tten

t /

%

% of Did Not Attend (DNA) Appointments - P Chart (Trial Limits)

20%

22%

24%

26%

28%

30%

32%

34%

36%

38%

Jan

-14

Feb

-14

Mar

-14

Ap

r-1

4

May

-14

Jun

-14

Jul-

14

Au

g-1

4

Sep

-14

Did

No

t A

tten

d /

%

% of Did Not Attend (DNA) Appointments - Line chart

33.924%

20%

22%

24%

26%

28%

30%

32%

34%

36%

38%

Jan

-14

Feb

-14

Mar

-14

Ap

r-14

May

-14

Jun

-14

Jul-

14

Au

g-14

Sep

-14

Oct

-14

Did

No

t A

tten

d /

%

% of Did Not Attend (DNA) Appointments - Run Chart

APPENDIX 5 – Chart Example 2 (Classification Data)

Page 18Contents Page

Type of data

Variables dataAttributes data

Multiple Observations

Single Observation

ClassificationCount

Here is an example of the SPC charts you’d expect when measuring classification data i.e. DNA %

Month DNA Total Appts % DNA

Jan-14 351 963 36.45%Feb-14 328 932 35.19%Mar-14 268 826 32.45%Apr-14 272 803 33.87%May-14 338 912 37.06%Jun-14 277 862 32.13%Jul-14 303 877 34.55%

Aug-14 212 624 33.97%Sep-14 271 817 33.17%Oct-14 300 976 30.74%Nov-14 258 903 28.57%Dec-14 256 816 31.37%Jan-15 247 917 26.94%Feb-15 281 952 29.52%Mar-15 291 906 32.12%Apr-15 350 942 37.15%May-15 249 741 33.60%Jun-15 286 845 33.85%Jul-15 251 960 26.15%

Aug-15 201 830 24.22%Sep-15 193 770 25.06%Oct-15 243 918 26.47%Nov-15 193 903 21.37%Dec-15 222 985 22.54%

Produced by Forid Alom

Page 21: How to use Statistical Process Control (SPC) charts?...Why do we use statistical process control (SPC) charts? Statistical Process Control (SPC) charts are used to study how a system

184.838

UCL276.678

LCL92.998

50

100

150

200

250

300

Jan

-14

Feb

-14

Mar

-14

Ap

r-14

May

-14

Jun

-14

Jul-

14

Au

g-14

Sep

-14

Oct

-14

No

v-14

Dec

-14

Jan

-15

Feb

-15

Mar

-15

Ap

r-15

May

-15

Jun

-15

Jul-

15

Au

g-15

Sep

-15

Oct

-15

No

v-15

Dec

-15

Jan

-16

Sick

nes

s /

Day

s

Days lost due to staff sickness - I Chart186.918

UCL309.561

LCL

64.275

50

100

150

200

250

300

350Ja

n-1

4

Feb

-14

Mar

-14

Ap

r-14

May

-14

Jun

-14

Jul-

14

Au

g-14

Sep

-14

Oct

-14

No

v-14

Dec

-14

Jan

-15

Feb

-15

Mar

-15

Sic

kness /

Days

Days lost due to staff sickness - I Chart (Trial Limits)

182.813

50

100

150

200

250

300

Jan

-14

Feb

-14

Mar

-14

Ap

r-14

May

-14

Jun

-14

Jul-

14

Au

g-14

Sep

-14

Oct

-14

Sick

nes

s /

Day

s

Days lost due to staff sickness - Run Chart

50

100

150

200

250

300

Jan

-14

Feb

-14

Mar

-14

Ap

r-1

4

May

-14

Jun

-14

Jul-

14

Au

g-1

4

Sep

-14

Sick

ne

ss /

Day

s

Days lost due to staff sickness - Line chart

APPENDIX 5 – Chart Example 3 (Variables Data)

Page 19Contents Page

Type of data

Variables dataAttributes data

Multiple Observations

Single Observation

ClassificationCount

Here is an example of the SPC charts you’d expect when measuring variables data i.e. sickness days

Date Sickness (Days)01-Jan-14 252

01-Feb-14 108

01-Mar-14 120

01-Apr-14 130

01-May-14 175

01-Jun-14 183

01-Jul-14 230

01-Aug-14 220

01-Sep-14 183

01-Oct-14 201

01-Nov-14 262

01-Dec-14 189

01-Jan-15 203

01-Feb-15 128

01-Mar-15 220

01-Apr-15 198

01-May-15 156

01-Jun-15 134

01-Jul-15 148

01-Aug-15 198

01-Sep-15 169

01-Oct-15 170

01-Nov-15 197

01-Dec-15 242

01-Jan-16 203

Produced by Forid Alom