Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
1
ILVO
Trend watching of microbiological results in the agro‐food processing chain
Koen De Reu and Lieve Herman
Symposium ‘Wat kunnen Nederlanders leren van de Belgen?Tuesday 26 February 2019
ILVO
Introduction
Trend watching Trend analysis: mathematical analysis of data series to detect statistically significant changes over time
Trend observation: data series are observed to discover visually possible evolutions
(based on EFSA 2010)
In agro‐food processing chain and in production process trend watching of microbiological and chemical analytical results
important tool to optimise internal quality management
detecting and identifying problems
quality and safety improvement
2
ILVO
Visiting a dozen companies in the agro‐food processing chain→ Examples throughout the chain
Trend watching of esp. microbiological analytical results
Trend watching examples detecting notable situations
Materials and methodsILVO
Trend watching in the primary production
3
ILVO
Not detected by management4 daysearlier
2 daysearlier
Real‐time predictive detectionTrend analysis
Problems – yes clear
Monitoring body weight laying hensCollected data benchmarked with expectations
Management notices infectious bronchitis virus 4 days after alarm
Only example of trend analysis (TA)
ILVO
Trend watching in the feed industry
4
ILVO
Trend observation in feed industry
Addition of propionic acid
ILVO
Trend watching in the food ingredient industry
a) Semi‐finished products
b) End products
5
ILVO
Heat treated semi‐finished product in the ingredient industry
• Confectionary applications, stabilizers, emulsifiers, coating agent in dragees, …
• High heat treatment followed by pumping, concentra on, filtra on, drying, … → semi‐finished product
• After this process, each batch control on ABSENCEof:
‐ Thermophilic spores (forming bacteria)
‐ Enterobacteriaceae‐ E. coli‐ Coliforms
ILVO
Heat treated semi‐finished product in the ingredient industry
50
60
70
80
90
100
110
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12
2014 2015
Success
rate ABSENCE(%
)
Success rate for ABSENCE total coliforms (10g) and thermophilic spores (2 x 30 g) on semi‐finished product
Thermofiles Coliforms
2018
Investigation in April 2018: Possible contamination source after heat treatment?→ Insufficient cleaning and disinfec on (CIP) of a newly and other type of pump
2017
6
ILVO
Trend observation between plantsSuccess rate ABSENCE coliforms on a semi‐finished product per plant
2017‐2018
60
70
80
90
100
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P1 P2 P3 P4 P5 P6 P7
Plant 1
Plant 2
Plant 3
Plant 4
60
70
80
90
100
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P1 P2 P3 P4 P5 P6 P7
Plant 4
Plant 5
Plant 6
60
70
80
90
100
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P1 P2 P3 P4 P5 P6 P7 P8
Plant 7
Plant 8
Plant 9
Plant 10
Trend observation (TO) between plants
ILVO
Trend observation in egg product industry
• Small middle sized enterprise
• Pasteurized liquid egg products (end prod.)
• Containers – Food industry (e.g. 1000l) Aseptic filling
• Microbiology: E.g. Total aerobic flora
– Target value < 1000 cfu/g (=spec)
< 1000 cfu/g From 1000 ‐ 10.000 cfu/g >10.000 cfu/g
Target Less than 10% batches per month Unacceptable
7
ILVO
MicrobiologyPercentage of samples in each product group above the spec max <10% (from 1.000 – 10.000 cfu/ml) KPI
above spec 2017 2018
Dec Jan Feb March April May
g Y salt 10% 0 1 0 0 4,58 0,5
g Y salt 11% 0 0,6 0 0 4,43 2,2
g Y salt 9% 0 ‐ 0 0 0 0
g Y salt 7% 0 0 0 0 0 0
g Y 0 0 0 0 0 0
g Y LTHT 0 0 0 0 0 0
g W 0 0,2 0 0 0 0,7
g W LTHT 0 10 0 0 0 0
g W salt 10% 0 0 ‐ 0 0 0
g EW 0 4,1 0 0 0 0
g EW LTHT 0 0 0 0 0 0
g Scrambled eggs 0 0 0 0 0 0
g RM seasoned 0 0 0 0 0 0
Action undertaken +follow up next months
Trend observation in egg product industry
ILVO
Trend watching in the food industry
a) On ingredients and end products
b) Environmental sampling after C&D (including equipment)
8
ILVO
Trend observation on purchased pasteurized milk by a dairy company
Supplier is informed and has to take
corrective actionsTO on supplied products
Total aerobic flora @ 37°C
Tank samples
Months
Number of analysis per month: (example delivered by Lavetan)
ILVO
Microbiological quality parameters in a sauce producing plant
• 14 sauce types randomly produced on 11 different production lines
• Self‐checking system: each batch microbiology
• Specifications (specs) on end products:– Total aerobic flora (TAF): ≤ 1000 cfu/g
– Lactic acid bacteria (LAB): ≤ 10 cfu/g
– Yeast: ≤ 100 cfu/g
– Moulds: ≤ 100 cfu/g
• The trend observations in % batches exceeding specifications using graphical representations
9
ILVO
Trend observations sauce plant 2018
0,0
1,0
2,0
3,0
4,0
5,0
6,0
7,0
8,0
9,0
TNT
AL 01
AL 02
AL 02 + 03
AL03
AL 04
AL 05
AL 08
AL 09
AL 11
AL 12
1e quarter (%)
% (TAF)
% (LAB)
% (yeasts)
% (moulds)
More exceeding of specs on the different lines in 1e quarter 2018
Almost no exceeding of specs on the different lines in 3e quarter 2018
Type 1, 2 and 3: TO in time, between parameters, between production lines
ILVO
0,0
1,0
2,0
3,0
4,0
5,0
6,0
7,0
8,0
9,0
TNT
AL 01
AL 02
AL 02 + 03
AL03
AL 04
AL 05
AL 08
AL 09
AL 11
AL 12
4e quarter 2017 (%)
% (TAF)
% (LAB)
% (yeasts)
% (moulds)
0,0
1,0
2,0
3,0
4,0
5,0
6,0
7,0
8,0
9,0
TNT
AL 01
AL 02
AL 02 + 03
AL03
AL 04
AL 05
AL 07
AL 08
AL 09
AL 11
AL 12
2e quarter 2018 (%)
% (TAF)
% (LAB)
% (yeasts)
% (moulds)
More mould problems 4e quarter 2017
Trend observations sauce plant 2017 ‐ 2018
More total flora problems 2e quarter 2018
Type 1, 2 and 3: TO in time, between parameters, between production lines
10
ILVO
Trend watching per product typeFilling line Recipe Exceeding of specs
75 76 141 182 208 211 212
TNT
AL 01 4AL 02
AL 02 + 03 1 5AL03 1 1AL 04 3
AL 05
AL 08
AL 09
AL 11 3 2 2
AL 12
totaal 1 1 12 1 3 2 2
Microbiological trend observation per product (=sauce) type
Type 4: TO on product type (cause: ingredient)
ILVO
Yeast contamination of dairy end products per prod. line
Production line 1
% of yeast contaminations = green bars
Production line 11
Trend observations in dairy industry
% Contaminated
sam
ples
% Contaminated
sam
ples TO between production lines
Date of production YY/MM Date of production YY/MM
11
ILVO
Trend observation of yeast contamination in dairy on production line 1
Start of production 3 hours samples later in production% Contaminated
sam
ples
% Contaminated
sam
ples
Cause:‐ Cross‐contamination from drain after C&D (CIP)‐ Specific line constructions (e.g. line 1)
TO during production% of yeast contaminations = green bars
Date of production YY/MM
ILVO
Trend of incidents for one dairy production zone (Lines 7‐8‐9‐10) over 3 months
Aroma dosage
Microbiological deterioration = yeasts
TO on incidents
12
ILVO
Trend observations on moulds in cheese industry
a) Trend in complaints
External complaints moulds on end products 2011 ‐ 2014 Number
ILVO
Trend observations on moulds in cheese industry
b) Trend in air contaminations
80,2
32,7
94,4
69
63,8
13,7
68,5
49
30,5
4,5
38,2
24
9,585,41
22,16
12
5,5 5,268,3
6
0
10
20
30
40
50
60
70
80
90
100
RDC 1e etage 2e etage average
CF
U/1
5 m
in
2010 2011 2012 2013 2014
Area 2 Area 3 Average
Moulds in the air in CFU/15 min
Corrective actions: 1) Correct air flow, 2) Fresh air cabins, 3) Intensive monitoring, …
Area 1
Type 1 and 2: TO in time and between areas
13
ILVO
Trend observations on moulds in cheese industry
c) Trend in air contaminations first floor
ILVO
Trend observations in meat industry
8%9%
7%
3%5%
6%
3%
7%
4% 3%2%
6%
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
% Not compliant microbiol. results: Plant 2 type product 1
10%
3%
0% 0% 0% 0% 0% 0% 0% 0%2%
0%0%2%4%6%8%
10%12%14%16%18%20%
% Non compliant microbiol. results: Plant 2 type product 2
Type 1 and 2: TO between products and in time
14
ILVO
Trend watching on environmental sampling in the food industry
(after cleaning and disinfection)
ILVO
Environmental hygiene monitoring after cleaning and disinfection in meat processing plant
Low care zone High care zone
Hygiene score
15
ILVO
3,44,2
0,7
3,2 3,1 2,8
0,8 0,82,1
1,1
3,0 2,8
12,2 11,9
18,7 18,4
26,2
17,0
10,09,4
4,63,1
5,3
12,0
7,8 8,1
9,710,8
14,6
9,9
5,4 5,1
3,32,1
4,1
7,4
13,6 13,6 13,6 13,6 13,6
10,1 10,1 10,1 10,1 10,1 10,1 10,1
0,0
5,0
10,0
15,0
20,0
25,0
30,0
35,0
aug‐17 sep‐17 okt‐17 nov‐17 dec‐17 jan‐18 feb‐18 mrt‐18 apr‐18 mei‐18 jun‐18 jul‐18
% non conform
results hygiene
% non conforming results hygiene in meat processing plants
Plant 1
Plant 2
Total
Objective
Parameters:Total aerobic floraEnterobacteriaceaeListeria spp.
Environmental hygiene monitoring after cleaning and disinfection trend observation between plants
Type 1 and 2: TO in hygiene in time and between plants
ILVO
Wholesale and cutting of meat
SME
Hygiene monitoring after
C&D
August 2014Defect dosing
pump disinfectant
Date sampling
Sample nr
ID place Area Sampling place Tot. flora
(cfu plate)
29-07-14 1 6 Snijzaal Snijplank konijn 229-07-14 2 14 Snijzaal Handvat deur met gang 4929-07-14 3 21 Fileerlijn Band bocht 029-07-14 4 29 Fileerlijn Pekeltank 029-07-14 5 38 Bereidingen Kapmachine 029-07-14 6 45 Bereidingen Worstendraaier 029-07-14 7 51 Algemeen Mes bereidingen 029-07-14 8 66 Algemeen Inpakmachine Yang groot 2629-07-14 9 71 Algemeen Inoxen tafel verpakkingsruimte 029-07-14 10 75 Algemeen Schreper plastiek 013-08-14 1 9 Snijzaal Snijmachine 2313-08-14 2 13 Snijzaal Prikblok 913-08-14 3 22 Fileerlijn Band flipper 16313-08-14 4 32 Fileerlijn Richtmessen aan band 313-08-14 5 39 Bereidingen Snijtafel 013-08-14 6 41 Bereidingen Kruidenweegschaal 4413-08-14 7 49 Algemeen Mes rood vlees 013-08-14 8 54 Algemeen Grijs bakje 313-08-14 9 58 Algemeen Handvat deur frigo gecal. filet 9413-08-14 10 63 Algemeen Hangrekken carré's 6619-08-14 1 Fileerlijn Borstkap >>19-08-14 2 Fileerlijn Band boven plankjes >>19-08-14 3 Snijzaal Snijtafel gevogelte 020-08-14 4 Fileerlijn Borstkap 16220-08-14 5 Fileerlijn Band boven plankjes 1921-08-14 6 Fileerlijn Borstkap 5921-08-14 7 Fileerlijn Band boven plankjes 022-08-14 8 Fileerlijn Borstkap >>22-08-14 9 Fileerlijn Band boven plankjes 010-09-14 1 Bereidingen Gehaktmolen 1711-09-14 2 Bereidingen Gehaktmolen 212-09-14 3 Snijzaal Kotelettekapper 012-09-14 4 Fileerlijn Borstkap 516-09-14 5 Bereidingen Cutter 016-09-14 6 Bereidingen Menger 017-09-14 7 Snijzaal Limamachine 218-09-14 8 Bereidingen Kapmachine 019-09-14 9 Bereidingen Gehaktmolen 4
16
ILVO
Hygiene monitoring after C&DWholesale and cutting of meat
Trends in types of contaminated surfaces
0,00
0,50
1,00
1,50
2,00
2,50
3,00
3,50
4,00
Ban
d been
deren
Ban
d been
deren
Ban
d been
deren
Ban
d bocht
Ban
d bocht
Ban
d bocht
Ban
d boven plankjes
Ban
d boven plankjes
Ban
d boven plankjes
Ban
d flip
per
Ban
d flip
per
Ban
d flip
per
Borstkap
Borstkap
Flap
einde prikb
lok
Flap
einde prikb
lok
Flipper
Flipper
Flipper
Messen
ontbener
Messen
ontbener
Messen
ontbener
Messen
ontveller
Messen
ontveller
Messen
ontveller
Naalden
blok
Naalden
blok
Naalden
blok
Opsteektafel
Opsteektafel
Opsteektafel
Opvangb
ak pekel
Opvangb
ak pekel
Opvangb
ak pekel
Pekeltan
kPekeltan
kPekeltan
kRichtm
essen aan
ban
dRichtm
essen aan
ban
dRichtm
essen aan
ban
dSchop
Schop
Snijp
lankjes
Snijp
lankjes
Snijp
lankjes
Tafel
Tafel
Weegschaal
Weegschaal
Weegschaal
kve/cm²
Plaats bemonstering
FILLETING LINE/ 2018
Tot. Kiem (kve/cm²)
TO hygiene of type of surfaces after C&D
ILVO
Trend watching in the catering
17
ILVO
Bacterial parameters of buffet meals
Evaluation microbiological results from 10 different locations of catering companyParameters:‐ TAF, E. coli, B. cereus, sulfite‐reducing
anaerobic bacteria
• = good• = acceptable• = unacceptable
Type 1 and 2: TO between locations and in time
2016
2017
(example delivered by Lavetan)
ILVO
Trend watching tools by service labs
(example delivered by Biotox)
18
ILVO
Package assessment
Half year overview meat factory
All sample types analysis packages
6,7%
93,3%
Beef hamburger10,7%
89,3%
ILVO
Most deviating matrix beef hamburger
19
ILVO
Deviating parameterILVO
Deviating parameter
20
ILVO
Take home messages
• In every stage of AFPC TW found to control microbiological quality and safety
• TW with wide variety of types of data handling
• Mainly trend observation is used
• Used to compare analytical results:• Products, parameters, batches, lines, area, plants, during production, time evolutions, incidents, complaints, …
• In all companies TW is used to:• Detect problems
• Improvement of quality and safety of products
→ TW very useful tool giving extra information out of analytical results
ILVO
11 invited (international speakers)
8 short communications
At least 40 poster presentations
300 participants
Fantastic social event
Cheap participation: 110 euro/1 day – 180 euro/2 days
More information: www.bsfm.be
21
ILVO
Words of thanks
Different concerned companies of the AFPC
Kristof Mertens (Prophyrio)
Jan Robrechts en Raf Peeters (Lavetan)
Sophie Sleuyter en Tom Benijts (Biotox/ECCA)
Saskia Desnyder (LIEMAQS bvba)
ILVO
Thank you
Research Institute for Agriculture,Fisheries and Food
Brusselsesteenweg 3709090 Melle – BelgiumT + 32 (0)9 272 30 00F +32 (0)9 272 30 01