PT Schemes from Colombia

Camilo DAleman C., Eliana Chavarro M., Immer Caicedo G.

Mol Labs Ltda, Bogot, Colombia
A matching exercise between proficiency-testing (PT) schemes from Colombia to Latin America, and global recognized schemes was done. We use
information from five schemes under ISO 17043:2010 and ISO 13528:2005, as we do. Comparison to 21 measurand at similar levels with more than 15
participants was done; we found small differences in RSD, as the use of robust statistics merits.

16,00
14,00
12,00
10,00
8,00
mg Na/L
pH units
mg SO4-2/L
mg Ca/L
mg Cl-/L
6,00
14,0% 15,0% 4,00
9,31 6,2% 257,0%
5030,0%
12015,0%
12,0%
109,8 10
27,1% 55,7
45,3
21,6
51,7
10,0%
6,0%
7,68
100
10,0%
2,00
20
8
4025,0%
10,5%
10,0%
10,0%
78,1
6,43 5,0%
31
29,5
10,0% 0,00
10,0%
80
8,0%
20,0%
154,0%
6
30
5,3%
60
6,0%2,6%
15,0% 6,3%
103,0%6,8 5,1%
4
20
36,7
3,3%
5,0%
405,0%
4,0%
10,0%
5,7
2,0%
9,21 5,1% 3,5% 10
5
2
0,7%
20
2,0%
1,0%
5,0%
0,0%
00,0%
0
00,0%
00,0%
0,0%
L
I
M
L
I
M
L
I
M
L
I
M
L
I
M
Assigned Value RSD
L - LEAP I - IDEAM M - Mol

Fig 1. PTs in drinking water: assigned value in columns, is a reference for measurand
level in matrix. On RSD lines: LEAP (L) assign perception values with a trend to
10%; IDEAM (I) Schemes, based on RTC certified reference materials, shows lower
RSD in all the exercises. Mol Labs (M) consensus values has high dispersion.
20
17,5%
44,3% 650,0%
12,0%
16,8
5,3
1,15
10,0%
100014,0%
1,06
40010,0% 13,2% 358,6
540,0%
789 15 12,0%
10,0%
12,6%
4,02
350 10,0%
12,0%
10,0%
8,5%
0,839
800
430,0%
3008,0%
10,0%
7,4%
10,0%
8,0%
5,7%250
10
7,3% 6008,0%
3
6,0%
6,0%
200 144
20,0%
6,1
6,0%
3,1%
400
2
1504,0%
10,0%
4,0%
5
3,55
4,0%
0,836
1002,0%
5,1%
200
110,0%
2,0%
29,4
2,0%88 36,8
50
00,0%
0 0,0%
00,0%
00,0%
L
I
M
L
I
M
L
I
M
L
I
M
L
I
M
Assigned Value RSD
L - LEAP I - IDEAM M - Mol

20,0%
15,0%
10,0%

0,0%

4
3
2
1
0
-1
-2
-3
-4

0
0,0%

0
B

0,5%
50
0,0%
0

B - BIPEA

5,0%
0,0%

M
Assigned Value

M - Mol Labs

RSD

Fig 3. Vegetable oil PTs: BIPEA (B) and Mol Labs (M). Consensus values on RSD.
Low differences between far away exercises. Mol Labs FAMES exercise has few (less
than 15) participants.
Moisture as g/100g
70
60

59,62

63,06

50
40
30
20
10

0,9%

1,1%

0
F

Ash as g/100g

gN/100g

Total Fat as g/100g

20,0%
4
3,50
2,93
3,5
3,7%
15,0%
3,4%
3
2,5
10,0%
2
1,5
5,0%
1
0,5
0,0%
0
F
M
F - FAPAS M - Mol Labs

18
5,0%
16
14
4,0%
12
10
3,0%
8
2,0%
6
4
1,0%
2
0
0,0%

15,96

7,5%
13,0

20,0%

20
9,0%
16,29*

7,0%
15

15,0%

5,0%
10
3,0%

4,0%

1,0%5
F
M
Assigned Value

10,0%
3,51
1,8%

5,3%

-1,0%0
M

RSD

Fig 4. PT in canned meat: FAPAS (F) perception values and Mol Labs (M) consensus.
Not so different values at > 17 participants in every measurand. But, Nitrogen measure
is out of set.
Total glycerin as g/100g

Density as g/mL
1

0,88537

0,8751

0,8

0,2%0,15

0,6

0,1% 0,1

0,4
0,2

0,2% 0,2

0,024%

0,125
17,6%
13,0%

0,1%0,05
0,011%

0
ASTM

0,17

0,0% 0

Mol Labs
A - ASTM

ASTM
M - Mol

Mol Labs

0,0623

50,0%
0,04

6,8%

Assigned Value

105,66 NR

115,20 2,46

115,20 0,28

177 PT schemes from Colombia (2008-2014) had a limit

on participant laboratories: some proposed exercises dont
arrive to a minimum of ten. Schemes for water and food
with measurand levels between g/L and mg/L are
successful, but for lower levels participants are not
enough for statistical purposes.
There are hardware predicaments by old equipment
everywhere, and scarce competences to the use of
significant figures, and to report uncertainty in
measurement data.
Bibliography

60,0%

[2] LEAP. Scheme Report EFF025. Effluent & Waste Water. May 2011.

50,0%

[3] Bureau InterProfessionnel d'Etudes Analytiques France. Rapport de

40,0%

Comparaisons Interlaboratoires. Huile d'olive 02-0121. Septembre 2010.

30,0%

[4] FAPAS. Report 018. Nutritional components in canned meat. October 2012.

20,0%

[5] ASTM. Committee D-2 Biodiesel Fuels Sample ID: BIOD1108. August 2011

10,0%
0,0%

ASTM

Fig 8. Uncertainty estimation is not the rule for this

laboratories, and there are strong differences within
reported uncertainty even though similar procedure and
equipment.

[1] LEAP. Scheme Report CHEM122. Potable Water. October 2013.

Water as g/100g
0,0700
40,0%
35,0%
0,0600
30,0%
0,0500
25,0%
0,0400
20,0%
0,0300
15,0%
0,0200
10,0%
0,0100
5,0%
0,0000
0,0%

5,0%
0,0%

2013

105,10 0,70

20,0%

10,0%
2,8%

2012

L201

1,9

150
1,0%
100

2011

Reported value and uncertainty mg Ca/L by SM 3500-Ca B

15,0%

10,8

2010

L200

192,6
1,6%

2009

L450

190,5

15,1
15,9%

-0,06

L188

19,1%

200
1,5%

2,5%
16
14
2,0%
12
10
1,5%
8
1,0%
6
4
0,5%
2
0
0,0%

2013

0,70 0,58

0,29

103,81 0,20

28,8%

1,2%

2,0%

0,48 0,73

L908

10

115,1

2,0%
250

2012

Fig 7. Consensus impact on performance evaluation: RSD

disperse values without trends on laboratory results.

100,59 8,11

12

11,8

140
50,0% 1,7%
120
40,0%
100
82,2
80
30,0%
60
20,0%
40
10,0%
20

2011

-0,6

L228

14

0,97

0,14

g FAME*/100 g

mg KOH/kg

mg I2/kg

2,0%

2,65

Fig 2. PTs in wastewater: LEAP (L) assign 10%, IDEAM RSD from reference
materials looks lower in every exercise, relative to the measurand level; Mol Labs (M)
consensus can be showing technical competence from the laboratories.
mEq O2/kg

5,3%

RSD

2009 2010 2011 2012 2014

5,0%

12,49

Historical data z-score

in Protein as g N/100g * 6,25
in canned meat

Historical data z-score

in Protein as g N/100g * 5,7
in wheat flour

MBAS mg/L

TSS as mg/L

2010

16,29

Fig 6. PTs in protein measurement along the years. RSD

value by statistical consensus: reasonable dispersion and
consistency.

106,80 NR

COD as mg O2/L

Assigned Value

L922

1,4
1,2
1
0,8
0,6
0,4
0,2
0

mg Cu/L

2009

2014

107,00 8,45

mg Cd/L

2013

L765

2012

109,00 NR

10

2011

L884

20

2010

106,94 2,44

30

2009

L46

40

18
10,0%
16,31
13,4%
8,3% 16
13,40
8,0%12,57
14
11,01
10,30 9,60 12
9,2%
9,25 5,5%
6,0%
5,0%
10
8
3,7%
4,0%
5,51
6
2,4%
3,4%
4
2,0%
2
0
0,0%

104,21 NR

50

9,4%
13,68

L101

60

Protein as g N/100g *6,25 in canned meat

Protein as g N/100g *5,7 in wheat flour

ISO Standard 13528:2005 defines five procedures to establish an assigned value and
another five to estimate the RSD for the measurand to be used in a PT evaluation. The
selection makes a difference between providers, but matrix and measurement level also
contributes.

Mol Labs
RSD

Fig 5. PT in Biodiesel Fuels: ASTM (A) and Mol Labs. Less significant figures and
high RSD for measurements from Colombia. Technological (equipment) difference?.

16,0%
14,0%
12,0%
10,0%
8,0%
6,0%
4,0%
2,0%
0,0%