Milk Quality Tracking and Tracing System as a Basis for Quality Based Milk Payment System

Milk Quality Tracking and Tracing System as
a Basis for Quality Based Milk Payment
System
A review of the project milestones and
proposed interventions
Ndungu Teresiah Wangui
MQT&T Project Manager
projectmanager@happycowkenya.com
FOR:
HAPPY COW LTD
28 September 2015 Ndungu T.W. 1

Introduction
• In Kenya, raw milk safety has been disputed over
a decade but no documented data exists.
• Greatest milk quality challenge for the CBE’s are
proper ways to maintain cold collection.
• The regulatory institutions are constrained
financially hence raw milk quality standards are
reluctantly applied and enforced.
• This creates a vacuum for farmers to switch
between buyers of raw milk specially due to
quality issues.
9/28/2015 Ndungu T.W. 2

Introduction
• The critical quality control challenges in line
with raw milk collection and bulking are;
adulteration, high bacterial load, high somatic
cell count and antibiotic residues.
• It therefore often fails the KEBS raw cow milk
specification on some of the parameters.
• Thus production of high quality products
becomes a challenge, financial losses incurred
and export potential limited.
9/28/2015 Ndungu T.W. 3

Introduction
HAPPY COW LTD
• A private dairy processing company
located in Nakuru, Kenya.
• It produces a wide range of dairy
products, sold in all major
supermarkets chains in Kenya and also
for exports to neighboring countries.
• The source of raw milk are CBE’s
namely Ngorika and Olenguruone
among many but the MQT&T project
will involve the two as a pilot.
• Historical data from HCL suggests
major challenge in TBC, CC,
adulteration and antibiotic residue that
are way above the KEBS standards.
9/28/2015 Ndungu T.W. 4

Historical data from an accredited lab – SOME EXTREME CASES
CBE A was supplying warm milk during this time.
9/28/2015 Ndungu T.W. 5
DATE CBE TBC/ML CC/ML RESAZURIN
TEST
ANTIBIOTIC
KEBS Version 2007 < 2,000,000 < 50,000 4-6 < 4 PPB
1 7/6/05 BULK SILO 98,000,000 9,900,000 4
2 2/2/10 BULK SILO 80,000,000 2,900,000 4
3 26/9/11 BULK SILO 160,000,000 1,900,000 4
4 9/7/14 CBE A 170,000,000 6,600,000 4
5 24/10/14 CBE A 89,000,000 7,900,000 4 POSITIVE
6 1/11/14 CBE B 190,000,000 2,000,000 5 POSITIVE
7 19/11/14 CBE A 110,000,000 1,700,000 5 POSITIVE
8 26/11/14 CBE B 10,000,000 190,000 4 POSITIVE
9 7/01/15 CBE A 110,000,000 2,700,000 4
10 28/01/15 CBE A 160,000,000 3,100,000 4
11 30/04/15 CBE A 100,000,000 1,400,000 4

Introduction
• Following an application to SNV/KMDP for
support in implementation of a QBMPs which was
accepted, a number of steps and investments along
the value chain were followed.
• The first step was the Zero Setting to be followed
by interventions, MQT&T system as a
precondition for QBMPs.
• The potential success of this pilot project, in the
smallholder supply chain, can likely be adopted by
other processors.
9/28/2015 Ndungu T.W. 6

Milk Quality Tracking & Tracing System
(MQT&T)
MQT&T will allow;
• Actors in the CBE’s identify the raw milk
quality at critical points in the collection
chain.
• The identification of the cans and groups.
• Identification of the causes contributing to
quality losses.
• Appropriate corrective actions.
Therefore, it must be able to connect the milk
quality in a can to a certain group of suppliers
for appropriate payment of their milk.
9/28/2015 Ndungu T.W. 7

MQT&T cont…....
• The poor infrastructure can lead to delays in
the collection chain for up to 6-7 hours
before milk processing.
• Therefore a drastic change in the way milk
is collected must be invented together with
the CBE’s.
• After success of MQT&T system, the
possibility of a QMBPs will be evaluated
and possibly implemented by year 2016.
9/28/2015 Ndungu T.W. 8

QUALITY BASED MILK PAYMENT SYSTEM
(QBMPS)
• The KDB has often advised the stake holders in
dairy sector to take responsibility and pilot a
QBMPS.
• QBMPS validates the quality and revenue of
raw milk delivered.
• Currently, increase in volumes and market share
are prioritized than milk quality assurance due
to vital factors as competition.
9/28/2015 Ndungu T.W. 9

QBMPS Practice
• QBMPS can instill responsibility on the farmers to
ensure delivery of quality raw milk hence
implementation will be with consent of the farmers.
• Basically the KISS approach (Keep it Simple and
Smart) should be applied in QBMPS .
• Adequate milk testing facilities should be provided and
penalties for added water, high bacterial load and
antibiotic positive milk.
• It should involve everyone’s efforts in the dairy value
chain (farmers, transporters, CBE and processors).
9/28/2015 Ndungu T.W. 10

Materials And Methods
The structure of the CBE’s
9/28/2015 Ndungu T.W. 11
SAMPLING
LEVELS
CBE A CBE B
Tanker 1 1
Cooling tanks 1 2
Collection routes 13 31
Project routes 6 6
Farmers cans 44 24

Materials and Methods
Sampling procedure
• Sampling levels were four;
• Composite can, composite route, cooler
immediately after it got filled and tanker
after its arrival at the processing plant.
NB: Samples for morning milk were taken at
random in the farmer’s container before
bulking.
9/28/2015 Ndungu T.W. 12

Materials And Methods
Study design
• The laboratory experimentation employed a
randomized complete block design with a 2 × 5
factorial arrangement and three replications.
• ANOVA was used to analyze the laboratory
results using PROC GLM procedure of the
statistical analysis system version 9.0 (SAS,
2006) and means separated using LSD
whenever there is variability at P ≤ 0.05.
9/28/2015 Ndungu T.W. 13

Statistical model for laboratory analysis;
𝒀𝒊𝒋𝒌𝒍 = 𝛍 + 𝛂𝐢 + 𝛃𝐣 + 𝜸 𝒌 + 𝜷𝜸𝒋𝒌 +Ɛ𝐢𝐣𝐤𝒍
where;
• Yijk is the response variable for milk quality.
• μ is the overall mean.
• αi is the ith replication effect.
• βj is the jth effect due to CBE blocking effect.
• γk is the kth effect due to treatment.
• βγjk is the jth CBE blocking effect on the kth
treatment.
• Ɛijk is the random error term.
9/28/2015 Ndungu T.W. 14

• Chemical Analysis
The chemical tests for organoleptic, alcohol,
lactometer, sediment, pH, freezing point, acidity, 10
minutes Resazurin, peroxide, antimicrobial, CMT
and antibiotics done according to KEBS (2007) and
FAO (2009).
• Microbial analysis
Total Plate Count and Coliform Count done
according to AOAC (2005) methods 991.14 and
990.12 respectively using 3MTM petrifilms plates
without modifications.
9/28/2015 Ndungu T.W. 15

9/28/2015 Ndungu T.W. 16
The serial dilutions

1
3
9/28/2015 Ndungu T.W. 17
21 2
33
4

Microbial Analysis
• The results are for the project
routes only.
• Sampling was done from the
nearest radius to the cooling
plant and the cooler sampled
immediately it got filled (not
yet chilled to 50C).
• It was stored and transported
separately for analysis and
follow up.
9/28/2015 Ndungu T.W. 18
0
5
10
15
20
25
30
CAN ROUTE COOLER TANKER
MILLLIONS/ML
SAMPLING LEVELS
CBE B
TBC/ML
CC/ML

9/28/2015 Ndungu T.W. 19
0
2
4
6
8
10
12
14
TBC CC LA RT DENS SED
TBC/CCMILLIONS/ML
TESTS ANALYZED
CBE B ROUTE COMPARISON ANALYSIS
ROUTE 1
ROUTE 2
ROUTE 3
ROUTE 4
ROUTE 5
ROUTE 6

Individual farmers sampling results
Test Best Worst TMTC
i.e. ˃˃ 999 Millions.
Average countable
Minus TMTC
TBC/ML 25,000 (5%) 34.3% 6,600,000
CC/ML 0 (14.3%) 31.4% 3,050,000
RESAZURIN
TEST 6 5 6
DENSITY 1.029 1.027 1.028
NB:
Freezing point determination suggests that 28.57% farmers had added water
while 42.86% of the samples had antibiotics residues.
9/28/2015 Ndungu T.W. 20

Composite can results (individual delivery)
CBE B (project routes)
• TBC/ML – 7.66 X 106
• CC/ML – 6.13 X 104
• RT – 6
• LA – 0.14
CBE A (project routes)
• TBC/ML – 3.49 X 107
• CC/ML - 9.70 X105
• RT – 6
• LA – 0.14
9/28/2015 Ndungu T.W. 21

Physical chemical tests
0
1
2
3
4
5
6 CBE B
RT
LA
DEN
SED
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
CBE A
RT
LA
DEN
SED
9/28/2015 Ndungu T.W. 22

Antibiotic Residues
This was done using:
1. At HC the two standards tests from Delvo.
The first test has high detection power but
takes 3hrs while the second one is quick
(5min) and tests majorly for β-lactams.
2. The activity/sensitivity test was done at the
two CBE checking for inhibitory substances
(“3hrs yoghurt culture test”).
9/28/2015 Ndungu T.W. 23

Antibiotic
Residues results
9/28/2015 Ndungu T.W. 24
Negative
46%
Positive
54%
CBE B
Negative
65%
Positive
35%
CBE A

California Mastitis Test
BEFORE AFTER
9/28/2015 Ndungu T.W. 25

CMT RESULTS
800-5000
15%
150-500
55%
400-1000
25%
0-200
5%
CBE B
800-5000
13%
150-500
65%
400-1000
22%
0-200
0%
CBE A
9/28/2015 Ndungu T.W. 26

Water adulteration
• This was done using a cryoscope.
• The cryoscope suggested that CBE A and B
had 23.8% and 36.8% water adulteration
respectively.
• This indicates that the thermo lacto density
meter might not be accurate enough.
9/28/2015 Ndungu T.W. 27

CORRELATIONS
• The results revealed a significant positive
correlation at p≤ 0.05 between time and total
bacterial count at the route level.
• The correlation between total bacterial count
and Resazurin test was not significant at p≤
0.05.
• Its alarming that no correlation was found.
• The sediment test and total bacterial test had a
positive correlation at p≤ 0.05.
9/28/2015 Ndungu T.W. 28

Added observations
• Samples of milk which
organoleptically smelled
and looked like a mixture
of water and flour and an
acidity of 0.07 to 0.09 %
and 0.09 % butter fat.
• Some milk delivered with
cement bags being used to
fasten the lids.
• More than 50% of the
cans were rejected at the
reception.
9/28/2015 Ndungu T.W. 29

9/28/2015 Ndungu T.W. 30
A B
C
D
B

Use of plastic jerry cans
9/28/2015 Ndungu T.W. 31

9/28/2015 Ndungu T.W. 32
2
2
1 3

Recording system for QBMPS
This will be done according to proposed delivery
receipt;
• The CBE name, the individual can number, the
collection point number, the route name, and the
sampling dates respectively.
The parameters to be included are;
• Date, milking time, milk volume, producers group
identification numbers/ names, date of sampling,
results of sampling on quality tests done and the name
of the grader.
• A robust software will be pretested and developed.
9/28/2015 Ndungu T.W. 33

Quality Indicators for QBMPS
• Total bacterial count.
• Coliform count.
• Somatic cell count.
• Presence of antibiotics residues
• Adulteration
NB: The samples must conform with the
parameters in the following table before
microbial analysis.
9/28/2015 Ndungu T.W. 34

Quality Tests KEBS Limits
1 Organoleptic
Free from off-flavours, odours, and be white creamish
in colour. No foreign matter observed.
2 Temperature <5-8°C
3 SNF >8.25 Total solids
4 Butterfat
>3.5 11.75
5 pH 6.60 – 6.75
6
Titratable Acidity
(% lactic acid)
0.14 -0.18
7 Specific Gravity 1.028 – 1.032
8 Freezing point -0.525 to -0.565
9 Alcohol test Negative
10 Hydrogen peroxide Negative
11 Antibiotics Negative9/28/2015 Ndungu T.W. 35

9/28/2015 Ndungu T.W. 36
HC LABORATORY

9/28/2015 Ndungu T.W. 37
HC LABORATORY

IN THE PAST
• Annato colour was sometimes added to rejected
milk to avoid human consumption.
• Only aluminium cans were used and were
thoroughly cleaned in the processing factories.
• Thermization/pre-pasteurization of raw milk with
immediate cooling before transportation for
pasteurization and packaging at other locations.
What went wrong?? We need to learn from the
past to improve the future.
9/28/2015 Ndungu T.W. 38

Proposed Project Interventions
• The construction of lab at CBE levels.
• Removal of plastic jerry cans and introduction
of vigorous can washing system.
• Organizing farmers into groups at the
collection points for faster collection.
• Training graders on milk testing and grading.
• Training the farmers and all milk handlers on
raw milk hygienic code of practice.
• Penalties on all non conforming milk.
• Payment based on quality.
9/28/2015 Ndungu T.W. 39

Alternative Payment Methods world wide
Payment according to quantity of milk is the
most common form of payment. Additionally
milk can also be paid according to;
• Composition and hygienic quality.
• Fat % in milk.
• Fat % and SNF% (“solids”).
• Seasonal payment according to fat% and
SNF%.
9/28/2015 Ndungu T.W. 40

Benefits of Improved Raw Milk Quality
9/28/2015 Ndungu T.W. 41
• Exports market expanded.
• Prolonged product shelf life.
• Improved product quality.
• Guaranteed food safety.
• Lowered processing costs.
• Farmers improved income.

Countries Which Have mplemented QBMP
Systems
• USA and Europe fully implemented.
• New emerging and important dairy African
countries have started QBMPS like
Zimbabwe and Zambia.
• Partly also in China, India, Vietnam,
……etc…..
• See next slide Brazil: Milk, Vector of
Development, Rennes- France, May 2014.
9/28/2015 Ndungu T.W. 42

Acknowledgement
• HC directors: Mr. and Mrs. Oosterwijk.
• Egerton supervisors: Dr. Muliro and Dr.
Omwamba.
• External backstopper Mr. Jan.
• SNV team.
• Happy Cow Ltd. team.
• Both CBE’s teams.
Thanks for listening
9/28/2015 Ndungu T.W. 44

Milk Quality Tracking and Tracing System as a Basis for Quality Based Milk Payment System

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to Milk Quality Tracking and Tracing System as a Basis for Quality Based Milk Payment System (20)

More from African Dairy Conference and Exhibition (20)

Recently uploaded (20)

Milk Quality Tracking and Tracing System as a Basis for Quality Based Milk Payment System

Editor's Notes