胡言乱语

外包,CMM,Six Sigma,软件服务,软件开发,软件过程管理,软件质量,项目管理,企业管理

  博客园 :: 首页 :: 博问 :: 闪存 :: :: 联系 :: 订阅 订阅 :: 管理 ::

Control charts selection guidelines 



u Charts - 单位缺陷数控制图


http://www.sytsma.com/tqmtools/uchart.html
Attribute control charts arise when items are compared with some standard and then are classified as to whether they meet the standard or not. The control chart is used to determine if the rate of nonconforming product is stable and detect when a deviation from stability has occurred. The argument can be made that a LCL should not exist, since rates of nonconforming product outside the LCL is in fact a good thing; we WANT low rates of nonconforming product. However, if we treat these LCL violations as simply another search for an assignable cause, we may learn for the drop in nonconformities rate and be able to permanently improve the process.

The u Chart is used when it is not possible to have an inspection unit of a fixed size (e.g., 12 defects counted in one square foot), rather the number of nonconformities is per inspection unit where the inspection unit may not be exactly one square foot...it may be an intact panel or other object, different in sizethan exactly one square foot. When it is converted into a ratio per square foot, or some other measure, it may be controlled with a u chart. Notice that the number no longer has to be integer as with the c chart.

Steps in Constructing a u Chart

  1. Find the number of nonconformities, c(i) and the number of inspection units, n(i), in each sample i.
  2. Compute u(i)=c(i)/n(i)
  3. Determine the centerline of the u chart:

  4. The u chart has individual control limits for each subgroup i.

  5. Plot the centerline, ubar, the individual LCL's and UCL's, and the process measurements, u(i).
  6. Interpret the control chart.

Example:

Besterfield Example:                    
data is from Besterfield (1990): Quality Control  p. 185                 

                        Number          Nonconformities
Day     Number          Non-            Per 
        Inspected       Conformities    Unit
1       110             120             1.0909
2       82              94              1.1463
3       96              89              0.9271
4       115             162             1.4087
5       108             150             1.3889
6       56              82              1.4643
7       120             143             1.1917
8       98              134             1.3673
9       102             97              0.9510
10      115             145             1.2609
11      88              128             1.4545
12      71              83              1.1690
13      95              120             1.2632
14      103             116             1.1262
15      113             127             1.1239
16      85              92              1.0824
17      101             140             1.3861
18      42              60              1.4286
19      97              121             1.2474
20      92              108             1.1739
21      100             131             1.3100
22      115             119             1.0348
23      99              93              0.9394
24      57              88              1.5439
25      89              107             1.2022
26      101             105             1.0396
27      122             143             1.1721
28      105             132             1.2571
29      98              100             1.0204
30      48              60              1.2500


Calculations:                   

UBAR =  1.2005          
                        
UCL =   ubar + 3*sqrt(ubar/n(i))                
LCL =   ubar - 3*sqrt(ubar/n(i))                

Day     CL      UCL             LCL             Nonconformities/Unit
1       1.2005  1.513900448     0.887091405     1.09
2       1.2005  1.563485937     0.837505915     1.15
3       1.2005  1.535975424     0.865016429     0.93
4       1.2005  1.507011595     0.893980258     1.41
5       1.2005  1.51678903      0.884202823     1.39
6       1.2005  1.639741695     0.761250158     1.46
7       1.2005  1.500557911     0.900433942     1.19
8       1.2005  1.532534517     0.868457335     1.37
9       1.2005  1.525958845     0.875033008     0.95
10      1.2005  1.507011595     0.893980258     1.26
11      1.2005  1.550892833     0.850099019     1.45
12      1.2005  1.59059276      0.810399092     1.17
13      1.2005  1.537736483     0.86325537      1.26
14      1.2005  1.524375074     0.876616779     1.13
15      1.2005  1.509712226     0.891279627     1.12
16      1.2005  1.55702269      0.843969162     1.08
17      1.2005  1.527566079     0.873425774     1.39
18      1.2005  1.707693252     0.693298601     1.43
19      1.2005  1.534241668     0.866750185     1.25
20      1.2005  1.543190862     0.857800991     1.17
21      1.2005  1.529197361     0.871794491     1.31
22      1.2005  1.507011595     0.893980258     1.03
23      1.2005  1.530853298     0.870138554     0.94
24      1.2005  1.635871613     0.76512024      1.54
25      1.2005  1.548918751     0.852073102     1.20
26      1.2005  1.527566079     0.873425774     1.04
27      1.2005  1.498088223     0.90290363      1.17
28      1.2005  1.521275681     0.879716172     1.26
29      1.2005  1.532534517     0.868457335     1.02
30      1.2005  1.674935581     0.726056271     1.25



u - Chart:






c Charts - 缺陷数控制图


http://www.sytsma.com/tqmtools/cchart.html
Attribute control charts arise when items are compared with some standard and then are classified as to whether they meet the standard or not. The control chart is used to determine if the rate of nonconforming product is stable and detect when a deviation from stability has occurred. The argument can be made that a LCL should not exist, since rates of nonconforming product outside the LCL is in fact a good thing; we WANT low rates of nonconforming product. However, if we treat these LCL violations as simply another search for an assignable cause, we may learn for the drop in nonconformities rate and be able to permanently improve the process.

The c Chart measures the number of nonconformities per "unit" and is denoted by c. This "unit" is commonly referred to as an inspection unit and may be "per day" or "per square foot" of some other predetermined sensible rate.

Steps in Constructing a c Chart

  1. Determine cbar.

    There are k inspection units and c(i) is the number of nonconformities in the ith sample.

  2. Since the mean and variance of the underlying Poisson distribution are equal,

    Thus,

    and the UCL and LCL are:

  3. Plot the centerline cbar, the LCL and UCL, and the process measurements c(i).
  4. Interpret the control chart.

Example:

Farnum Example:				
data is from Farnum (1994): 				
Modern Statistical Quality Control and Improvement, p. 248				

	Non-conforming			
Day	Errors/1000 lines			
1	6			
2	7			
3	7			
4	6			
5	8			
6	6			
7	5			
8	8			
9	1			
10	6			
11	2			
12	5			
13	5			
14	4			
15	3			
16	3			
17	2			
18	0			
19	0			
20	1			
21	2			
22	5			
23	1			
24	7			
25	7			
26	1			
27	5			
28	5			
29	8			
30	8			
				
				
Calculations:				
				
CBAR =	4.4667			
				
UCL =	cbar + 3*sqrt(cbar) =	10.80701366	
LCL =	cbar - 3*sqrt(cbar) =	-1.873680327 = 0
	(when LCL < 0, set LCL = 0)	
		
Day	CL	UCL		LCL	NonConforming
1	4.4667	10.80701366	0	6
2	4.4667	10.80701366	0	7
3	4.4667	10.80701366	0	7
4	4.4667	10.80701366	0	6
5	4.4667	10.80701366	0	8
6	4.4667	10.80701366	0	6
7	4.4667	10.80701366	0	5
8	4.4667	10.80701366	0	8
9	4.4667	10.80701366	0	1
10	4.4667	10.80701366	0	6
11	4.4667	10.80701366	0	2
12	4.4667	10.80701366	0	5
13	4.4667	10.80701366	0	5
14	4.4667	10.80701366	0	4
15	4.4667	10.80701366	0	3
16	4.4667	10.80701366	0	3
17	4.4667	10.80701366	0	2
18	4.4667	10.80701366	0	0
19	4.4667	10.80701366	0	0
20	4.4667	10.80701366	0	1
21	4.4667	10.80701366	0	2
22	4.4667	10.80701366	0	5
23	4.4667	10.80701366	0	1
24	4.4667	10.80701366	0	7
25	4.4667	10.80701366	0	7
26	4.4667	10.80701366	0	1
27	4.4667	10.80701366	0	5
28	4.4667	10.80701366	0	5
29	4.4667	10.80701366	0	8
30	4.4667	10.80701366	0	8


c - Chart:







p Charts - 不合格率控制图


http://www.sytsma.com/tqmtools/pchart.html
Attribute control charts arise when items are compared with some standard and then are classified as to whether they meet the standard or not. The control chart is used to determine if the rate of nonconforming product is stable and detect when a deviation from stability has occurred. The argument can be made that a LCL should not exist, since rates of nonconforming product outside the LCL is in fact a good thing; we WANT low rates of nonconforming product. However, if we treat these LCL violations as simply another search for an assignable cause, we may learn for the drop in nonconformities rate and be able to permanently improve the process.

p Charts can be used when the subgroups are not of equal size. The np chart is used in the more limited case of equal subgroups.

Steps in Constructing a p Chart

  1. Determine the size of the subgroups needed. The size, n(i), has to be sufficiently large to have defects present in the subgroup most of the time. If we have some idea as to what the historical rate of nonconformance, p, is we can use the following formula to estimate the subgroup size:

    n=3/p

  2. Determine the rate of nonconformities in each subgroup by using:

    phat(i)=x(i)/n(i)

    where:

    phat(i)=the rate of nonconformities in subgroup i

    x(i)=the number of nonconformities in subgroup i

    n(i)= the size of subgroup i

  3. Find pbar; there are k subgroups.

  4. Estimate sigma-p if needed and determine the UCL and LCL:

     

  5. Plot the centerline, pbar, the LCL and UCL, and the process measurements, the phat's.
  6. Interpret the data to determine if the process is in control.

Example:

Farnum Example:					
data is from Farnum (1994): 					
Modern Statistical Quality Control and Improvement, p. 242					
		Number			
Day	Rejects	Tested	Proportion		
1	14	286	0.0490		
2	22	281	0.0783		
3	9	310	0.0290		
4	19	313	0.0607		
5	21	293	0.0717		
6	18	305	0.0590		
7	16	322	0.0497		
8	16	316	0.0506		
9	21	293	0.0717		
10	14	287	0.0488		
11	15	307	0.0489		
12	16	328	0.0488		
13	21	296	0.0709		
14	9	296	0.0304		
15	25	317	0.0789		
16	15	297	0.0505		
17	14	283	0.0495		
18	13	321	0.0405		
19	10	317	0.0315		
20	21	307	0.0684		
21	19	317	0.0599		
22	23	323	0.0712		
23	15	304	0.0493		
24	12	304	0.0395		
25	19	324	0.0586		
26	17	289	0.0588		
27	15	299	0.0502		
28	13	318	0.0409		
29	19	313	0.0607		
30	12	289	0.0415		


Calculations:					

PBAR =	0.0539				

UCL =	pbar + 3*sqrt(pbar*(1-pbar)/n(i))				

LCL =	pbar - 3*sqrt(pbar*(1-pbar)/n(i))				

Day	CL	UCL		LCL		Proportion
1	0.0539	0.093892049	0.013808661	0.0490
2	0.0539	0.094246721	0.013453989	0.0783
3	0.0539	0.092310827	0.015389883	0.0290
4	0.0539	0.092126068	0.015574642	0.0607
5	0.0539	0.093410843	0.014289867	0.0717
6	0.0539	0.092624795	0.015075915	0.0590
7	0.0539	0.091587368	0.016113342	0.0497
8	0.0539	0.091943946	0.015756764	0.0506
9	0.0539	0.093410843	0.014289867	0.0717
10	0.0539	0.093822229	0.013878481	0.0488
11	0.0539	0.092498288	0.015202422	0.0489
12	0.0539	0.091240619	0.016460091	0.0488
13	0.0539	0.093209857	0.014490853	0.0709
14	0.0539	0.093209857	0.014490853	0.0304
15	0.0539	0.091883814	0.015816896	0.0789
16	0.0539	0.09314354	0.01455717	0.0505
17	0.0539	0.094103724	0.013596986	0.0495
18	0.0539	0.091646103	0.016054607	0.0405
19	0.0539	0.091883814	0.015816896	0.0315
20	0.0539	0.092498288	0.015202422	0.0684
21	0.0539	0.091883814	0.015816896	0.0599
22	0.0539	0.091528906	0.016171804	0.0712
23	0.0539	0.092688517	0.015012193	0.0493
24	0.0539	0.092688517	0.015012193	0.0395
25	0.0539	0.091470715	0.016229995	0.0586
26	0.0539	0.093683678	0.014017032	0.0588
27	0.0539	0.093011904	0.014688806	0.0502
28	0.0539	0.091823966	0.015876744	0.0409
29	0.0539	0.092126068	0.015574642	0.0607
30	0.0539	0.093683678	0.014017032	0.0415


p - Chart:




np Charts - 不合格数控制图


http://www.sytsma.com/tqmtools/npchart.html
Attribute control charts arise when items are compared with some standard and then are classified as to whether they meet the standard or not. The control chart is used to determine if the rate of nonconforming product is stable and detect when a deviation from stability has occurred. The argument can be made that a LCL should not exist, since rates of nonconforming product outside the LCL is in fact a good thing; we WANT low rates of nonconforming product. However, if we treat these LCL violations as simply another search for an assignable cause, we may learn for the drop in nonconformities rate and be able to permanently improve the process.

The np Chart can be used for the special case when the subgroups are of equal size. Then it is not necessary to convert nonconforming counts into the proportions phat(i). Rather, one can directly plot the counts x(i) versus the subgroup number i.

Steps in Constructing an np Chart

  1. Determine the size of the subgroups needed. The size, n, has to be sufficiently large to have defects present in the subgroup most of the time. If we have some idea as to what the historical rate of nonconformance, p, is we can use the following formula to estimate the subgroup size:

    n=3/p

  2. Find find pbar.

  3. Find the UCL and LCL where:

  4. Plot the centerline pbar, the LCL and UCL, and the process nonconforming counts, the x(i)'s.
  5. Interpret the control chart. Only if a point is outside the +/- 3 sigma range is the process considered to be out of control.

Example:

Farnum Example:				
data is from Farnum (1994): 				
Modern Statistical Quality Control and Improvement, p. 245	
			
			Sample		
Day	Non-conforming	Size		
1	10		100		
2	12		100		
3	10		100		
4	11		100		
5	6		100		
6	7		100		
7	12		100		
8	10		100		
9	6		100		
10	11		100		
11	9		100		
12	14		100		
13	16		100		
14	21		100		
15	20		100		
16	12		100		
17	11		100		
18	6		100		
19	10		100		
20	10		100		
21	11		100		
22	11		100		
23	11		100		
24	6		100		
25	9		100		


Calculations:			

PBAR =	0.1088		
CL =	10.8800		

UCL =	n*pbar + 3*sqrt(n*pbar*(1-pbar))		

LCL =	n*pbar + 3*sqrt(n*pbar*(1-pbar))	
	

Day	CL	UCL		LCL		NonConforming
1	10.8800	20.22164354	1.538356462	10.0000
2	10.8800	20.22164354	1.538356462	12.0000
3	10.8800	20.22164354	1.538356462	10.0000
4	10.8800	20.22164354	1.538356462	11.0000
5	10.8800	20.22164354	1.538356462	6.0000
6	10.8800	20.22164354	1.538356462	7.0000
7	10.8800	20.22164354	1.538356462	12.0000
8	10.8800	20.22164354	1.538356462	10.0000
9	10.8800	20.22164354	1.538356462	6.0000
10	10.8800	20.22164354	1.538356462	11.0000
11	10.8800	20.22164354	1.538356462	9.0000
12	10.8800	20.22164354	1.538356462	14.0000
13	10.8800	20.22164354	1.538356462	16.0000
14	10.8800	20.22164354	1.538356462	21.0000
15	10.8800	20.22164354	1.538356462	20.0000
16	10.8800	20.22164354	1.538356462	12.0000
17	10.8800	20.22164354	1.538356462	11.0000
18	10.8800	20.22164354	1.538356462	6.0000
19	10.8800	20.22164354	1.538356462	10.0000
20	10.8800	20.22164354	1.538356462	10.0000
21	10.8800	20.22164354	1.538356462	11.0000
22	10.8800	20.22164354	1.538356462	11.0000
23	10.8800	20.22164354	1.538356462	11.0000
24	10.8800	20.22164354	1.538356462	6.0000
25	10.8800	20.22164354	1.538356462	9.0000



np - Chart:

posted on 2005-03-08 21:16  台风  阅读(2106)  评论(0编辑  收藏  举报