Content uploaded by Lukas Bertschinger
Author content
All content in this area was uploaded by Lukas Bertschinger on Dec 22, 2016
Content may be subject to copyright.
technical report
229
Innovative Concepts and Technologies for Supply Chain Quality
Management of Fruit and Vegetable and Improvement of
monitoring systems
Noga, G.1, Schulpin, C.1, van Kooten, O.2, Schouten, R.2, Bertschinger, L.3, Crespo, P.3
1 Bonn University, Institute for Crop Science and Resource Conservation, -Horticultural Science-, Auf dem Hügel 6, 53121
Bonn, Germany, E-mail: NogaG@uni-bonn.de
2 Wageningen University, Horticultural Production Chains, Marijkeweg 22, 6709 PG, Wageningen, The Netherlands
3 Agroscope Changins-Wädenswil Research Station ACW, P.O. Box 185, CH – 8820, Wädenswil, Switzerland
Keywords: modelling, product quality, rmness, colour, apples, tomatoes, carrots,
Chalara sp.
1. Introduction
At the end of the 20th century signicant changes have been introduced in food supply systems, which
contributed to a better fullment of consumers´ demands (Early 2005), as conrmed by enhanced
consumer willingness to buy the produce. With reference to perishable food, it denes the acceptance
period of the quality of fresh fruits and vegetables (Tijskens 1997 and 2000). In order to determine the
acceptance period of a product, appropriate quality attributes for product quality characterization, sui-
table quality measurement technologies and models for predicting product quality at each point within
the supply chain have to be elaborated and combined (Schouten and van Kooten 1998).
Quality management systems in supply and demand chains of fruit and vegetables are currently being
developed in several large scale projects in the EU (Snoekx et al. 2005). These projects describe the
relative quality decay in relation to environmental constraints during transport in the chain. However,
consumers are not interested in relative quality but absolute quality at the moment of purchase.
Therefore the challenge is to determine the actual quality of the product at any time within the chain.
This can be achieved if the initial quality of the product at the point of harvest is known and if this in-
formation can be used in combination with the environmental parameters within the chain. In order to
get an estimate on the initial quality we have to understand how the process of production leads up to
a nal quality in the growth phase, i.e. an initial quality in the post-harvest phase (Hertog et al. 2004),
which can be maintained at a constant level in the ideal case, but more or less will deteriorate in practi-
ce. Combining this initial quality with the information obtained from tracking the product in the chain is
rather straight-forward.
On the fresh produce sector European supply chains will have to compete more and more with overseas-
oriented ones regarding produce like apples, onions, melons and even carrots. In order to assure market
share and competitive ability of regional products they must show an advantage in maturity, freshness
and shelf life potential, recognizable for the consumer. In order to deliver transparency not only on the
safety level but also on the quality level, this effort will make each participant of such supply chains an
important member contributing to a consumer-oriented high quality produce.
One item that has been out of focus too much, however, is the need to understand how different regions
and different seasons do affect quality and quality behaviour. Differences evoked or caused by different
cultivars, growing sites, soil types, climate and weather etc. have to be merged and combined into
one description. For that type of understanding and integration a different approach is necessary that
incorporates the relevant behaviour of the product, both in the preharvest (food production) as in the
postharvest realm (distribution and processing). Traditionally approaches in modelling (mainly statistical
and/or empirical models) are unsuitable to accommodate this integration of knowledge (Tijskens et al.
2006). The ultimate goal of modelling is to predict future behaviour of any product, in any circumstance,
from any region and grown in any season.
Innofruit & QualiTools
technical report
230
Up to now the quality systems in use in the regions of application do not allow to predict how quality will
evolve along the chain towards the consumer. The reason is that the physiological and the phytosanitary
status of the product at harvest (initial quality) cannot be measured accurately, so far. A decay of quality
up to the point of sale is very often the consequence.
The aim of this project therefore was to overcome this problem by introducing appropriate measure-
ment techniques before/at harvest and fusing the measured data in such a way that a more accurate
prediction of product quality can be made. Moreover, a combination of diagnostic and prospective tools
that can be applied to estimate the absolute quality including the phytosanitary status of fruits and ve-
getables was to be exploited. As basis, an already existing model for predicting quality decline during
marketing of fruits and vegetables had to be validated rst. This model is based on the biological vari-
ance of quality and phytosanitary attributes of the product at harvest (Schouten et al. 1997). The vari-
ance will be determined, and the obtained results serve as a basis for the prediction of the forthcoming
quality status in the running supply chain, e.g. from the farm to the consumer (Tijskens et al. 2005).
output: absolute
product quality
model for predicting
product quality
input: initial product
qualität
Seed/
Plant material
Plantprotection
Nutrition
Producer Distributer Retailer
Concept for predicting product quality
at each point of the supply chain
Consumer
win-win-situation
Consumer
Chain partners
cultivation
data, biological variance, storage and transport data
Decision tool for all market partners
Figure 1: Concept for innovative technologies for supply chain quality management of fruit
and vegetables
Therefore, as a result of the project a novel quality assessment tool can be expected, which can be
used to predict quality at any point of the supply chain (Fig. 1). Consequently, the decision makers of
fresh produce trade in the Euregio will use that system to improve their efforts in supplying the different
markets with suitable products. Finally the consumer in the regions will be provided with a higher per-
centage of high-quality products, which also allow to extend shelf life after the purchase.
For evaluating innovative technologies and approaches to improve supply chain quality management of
fruit and vegetable it was of particular interest to assess data on
quality attributes that characterize the product quality (incl. pathogen contamination grade) - most appreciably
measurement technologies that are applicable, easy to handle and qualied to determine - changes in quality attributes of individual products and pathogen contamination grade of fresh
products
parameters with inuence on the product quality and pathogen contamination in the pre- and - post-harvest phase
combination of suitable, non-destructive measurement technologies in order to estimate -
representative quality attributes with specic quality models for prediction of product quality
Innofruit & QualiTools
technical report
231
2. Materials and methods
2.1 Plant material
2.1.1 Apples:
Malus domestica
Borkh. cvs. “Braeburn”, “Golden Delicious” and “Jonagold” cultivated under commercial
growing conditions at the Research Station Klein-Altendorf, Bonn University, North-Rhine Westphalia
(NRW). Different cultivars were selected to include fruits differing in skin colour, e.g. green-yellow (“Gol-
den D.”), red-green (“Braeburn”) and red-yellow (“Jonagold”) - because parameters to be evaluated are
inuenced by pigment content and composition.
2.2.2 Tomatoes:
Truss tomatoes
Lycopersicum esculentum cv
. “Cedrico RZ” cultivated at the Vegetable Research Station
Marhof, Bonn University (NRW) under protected cultivation and practice-related conditions.
2.2.3 Carrots:
Daucus carrota
L. cv. “Bolero” cultivated in farmers’ elds at different locations in Switzerland.
2.2 Technologies for measuring product quality attributes
The common but mostly labour-intensive methods used in practice to evaluate fruit quality and matu-
rity involve destructive estimation of fruit rmness, starch breakdown, analyses of soluble solids and
titratable acids content in esh (Streif, 1996). This is, however, labour and time consuming. Also, with
standard methods there is no possibility to monitor maturity or quality changes in the same individual
product. Therefore innovative measurement methods are required, such as non-destructive sensor tech-
nologies for rapid evaluation of maturity grade in order to estimate the optimal harvest time, for quality
detection of the fruits, for automation of commercial fruit grading and for monitoring or predicting fruit
quality and storage ability pre- and post-harvest.
2.2.1 Conventional methods used as references
The suitability of innovative non-destructive technologies for determining product quality needs to be
validated by comparison with standard reference methods (Kuckenberg et al., 2008). Comparative stu-
dies were conducted with “Golden Delicious” and “Jonagold” fruit, harvested in October 2006. Fruits
with a diameter of 75/80 mm were selected, assigned to 3 treatment groups and stored in the dark at
10°C, 15°C, and 20°C, respectively, for three weeks. For apple, quality attributes, such as rmness and
colour, were regarded as representative parameters for characterisation of fruit quality and were mea-
sured immediately after harvest as well as 10 and 20 days thereafter.
Firmness - Penetrometer
Firmness of fruits was measured with a handheld penetrometer (destructive technique) at the equatorial
level on two opposite sides of each fruit. With this method, the force needed to penetrate the fruit esh
with a stamp area of 1 cm2 is measured.
Colour - Chemical analysis
The chlorophyll contents (a, b and total chlorophyll) in the fruit skin were determined photometrically
after chemical extraction of the green pigments with dimethyl sulfoxide (DMSO, Blanke 1992).
2.2.2 Innovative non-destructive measurement techniques
Non-destructive technologies offer the opportunity to determine quality changes repeatedly in each
individual fruit. This aspect is of great importance for developing product quality models. However, the
Innofruit & QualiTools
technical report
232
technologies differ in their measurement principles, the detection parameters as well as sensitivity and
precision of measurements. Hence, innovative sensor technologies need approval of their acceptability
as evaluation techniques for selected quality parameters and have to be validated for individual crops
and cultivars because of potential differences in specic fruit skin colouration, fruit esh texture and
elasticity, as well as fruit size and shape.
Firmness - Intelligent Firmness Detector (IFD)
Several non-destructive sensor technologies for measuring rmness have been developed (Höhn and
Winkler 2003), e.g. acoustic technologies to evaluate the texture or pressure by measuring the elasticity
of the fruit skin (Courcoux et al. 2005).
In the present work, rmness of apple and tomato fruit was measured with the “Intelligent Firmness
Detector” (IFD, Greefa Co., Geldermalsen, Netherlands). The instrument employed was originally de-
signed as component of a sorting machine; currently it is not used for commercial maturity or quality
controls. The IFD rmness module records the fruit elasticity (IFD-Index) along the fruit equator 20
times within one rotation. Velocity of rotating roles has to be adjusted for each species, cultivar and
class of fruit size.
Colour
Ripening and senescence processes are accompanied by changes in the content and composition of
pigments in the skin and esh of fruits. Thereby the green colour resulting from the chlorophyll content
decreases and the red colour as a result of the anthocyanin content increases. In dependence from the
kind of crop and cultivar other pigments like carotenoids with orange colour are of specic interest for
evaluation of maturity or quality of products.
Pigment Analyzer (PA):
Colour was measured with a handheld pigment analyzer (PA, CP Co., Falkensee, Germany. The innova-
tive potential of this relatively new instrument is based on the fruit light remission measurement with
simultaneous detection of different pigments within one recording. For evaluation of chlorophyll content
(fruit ground color), remission intensities at 650 (R650) and 780 (R780) nm were estimated and Norma-
lized Differenced Vegetation Index calculated as NDVI = (R780-R650)/(R780-R650). Anthocyanin con-
tent was evaluated with the Normalized Anthocyanin Index (NAI value) by remission (R) measurement.
The respective formula is as follows: NAI = (R 780-R 570)/(R 570+ R 780). Before taking PA readings,
a circle with a diameter of 2 cm was marked, both on the sun and shade-exposed side of each individual
fruit. This was to follow-up colour changes even on the same spots on the fruit surface.
Mini-Veg N:
MiniVeg is a non-destructive instrument to detect chlorophyll content in plants and fruits by Laser In-
duced Fluorescence (LIF). The instrument has been developed by Fritzmeier Co., Großhelfendorf, Ger-
many, and is used for estimating the nitrogen supply as basis for site-specic fertilisation of agricultural
crops. In the present investigation the MiniVeg sensor was applied to test whether LIF is an appropriate
method to assess maturity grade and quality of fruits and vegetables (Kuckenberg et al., 2008).
The objective of these investigations was to document that chlorophyll degradation is closely related to
changes of maturity or quality attributes in apple fruit. The research studies were performed with apple
cultivars differing in colour like “Golden Delicious” (green-yellow) and “Jonagold” (red-yellow). Measu-
rements were taken over a period of 3 weeks with 8 readings at a 3 to 4-day interval both on the sun
and shade exposed fruit sides.
Innofruit & QualiTools
technical report
233
2.3 Batch quality models
Predicting product quality requires establishing a model on the dynamics of product quality changes.
In order to achieve this ultimate goal, detailed knowledge on the underlying basic processes is needed.
This can only be achieved by implementation of data into so-called fundamental, process-oriented mo-
dels. The new approach of the applied quality model is based on consideration of the biological age of
the products as a tool for modelling in pre- and postharvest horticulture. The biological age includes:
the biological time - the physiological time - the state of development -
and depends on various and varying factors during growing period and shelf life, e.g. fertilisation, di-
sease status, position on the tree and environmental conditions within the post-harvest phase (Hertog
2006, Tijskens and Konopacki 2003). These biological variation inuences the maturity stage at harvest
and senescence processes in the post-harvest and shelf life period (Hertog et al. 2002, 2004, Tijskens
et al. 2003). In combination with the calendar time it is expressed as the Biological Shift Factor (BSF,
Tijskens et al. 2005).
The biological shift factor can be estimated and data pooled. The factor is directly related to the biolo-
gical variation of batches, populations or individuals and an important factor in the new approach for
modelling product quality.
In the present work for apples and tomatoes, the effect
of maturity level at harvest, storage period and tempe-
rature on product quality of apples and tomatoes had
to be analyzed. Therefore batches were combined,
whereas a batch is dened as products with the same
growing history. Hence, the batches were considered
as a) colour and rmness behaviour at harvest (initial
quality) and b) the colour and rmness behaviour as
function of storage time and storage temperature. Sto-
rage of batches at different temperatures was neces-
sary to calibrate quality change models. A variation in
initial quality attributes was essential, as colour and rmness are linked per batch. In order to reach the
overall objective of predicting product quality, colour and non-destructive rmness data were included
into quality models, which may be applied to apple, tomato and other fruits or vegetables.
In the case of carrots a batch consisted of an amount of carrots that were wounded and inoculated with
Chalara sp
. spores at a determined level. The aim of the study was to determine the maximum tempe-
rature at which carrots can be stored without appearance of
Chalara
symptoms during distribution and
sales. To be able to predict the losses caused by
Chalara sp
., it is important to understand the disease
progress in fresh and stored carrots at different temperatures.
2.3.1 Apple batches
Fruits of the cv. “Braeburn” were harvested at three different maturity stages (unripe, ripe, overripe),
separately harvested from sun and shade-exposed sides and from three different positions within the
tree (top, middle, bottom). The apples consisted of 18 batches of 40 fruits each, and were stored in the
dark at three different temperatures (5°C, 15°C, 20°C) for three weeks after each harvest date. Firm-
ness and colour were measured at a constant interval depending on storage temperature. The combined
apple batches might be characterised as a function of position in the tree, sun/shade side and harvest
maturity (pre-harvest parameters) with regard to colour and rmness changes at different storage tem-
peratures (post-harvest treatments).
Variability of the
biological age
0246810
40
50
60
70
80
90
100
kleur (Hue)
tijd
harvest
Innofruit & QualiTools
technical report
234
2.3.2 Tomato batches
Fruits, cultivated at two different concentrations of Ca nutrition (4,5 mmol/l Ca, as commonly used in
practice, and by 30 % reduced Ca supply) were harvested on October 26, 2006, with a diameter of
about 40 mm in three different maturity stages (green = ”breaker stage”, orange = ”pink stage”, red =
”red stage”). Fifteen tomato batches were combined and stored in the dark at 3 different temperatures
(10°C, 15°C, 20°C) for 3 weeks. Firmness and colour were measured at a constant time interval. The
tomato batches might be characterised as function of Ca supply and harvest maturity grade with regard
to postharvest colour and rmness changes at different storage temperatures (Lana et al. 2005).
2.3.3 Carrot batches
Contamination of carrots with black rot fungi (
Chalara
sp.) are a tremendous problem in Switzerland
(Heller et al. 2005) and originates in the eld but post harvest factors like harvest time, storage duration
and conditions, washing and packing procedures and temperature inuence disease outbreak at the
retailer level. In most cases the symptoms do not develop at the moment of sorting and packing but
rather appear in the store or even after the consumer purchased the carrots. For a better quantitative
understanding of disease progress in time and for later disease progress modelling as a forecasting and
management tool, disease progress in time at different temperatures was measured in different carrot
batches characterised by determined inoculation levels.
Carrots were grown in seramis or in seramis soil mixtures which have been previously disinfected by
heat treatment. (>60°C). After the harvest carrots were stored at temperatures between 0°C and 2°C
and were used for the trial at three different times. The rst experiment started directly after harvest
(July/August 2006), the second one after two month of storage (Sept. 2006) and the third one after 5
month of storage (January 2007).
C. elegans
and
C. thielavioides
were used for the inoculation of the carrots, except for the second expe-
riment only with C. elegans. The conidial suspension [30*103 spores/ml] of
C. elegans
and
C. thielavi-
oides
was made from 7-10 day old culture, crowing on malt agar.
Each carrot was wounded at 3 spots with an acerb borer (8 mm diameter and at a depth of 1-2 mm): on
the top (top), in the center (medium) and near the root tip (peak). In every wound 10µl of the suspensi-
ons was inoculated. The carrots were incubated at 5 different temperatures: 1, 2, 4, 8, and 20°C, were
placed individually on moist paper towels in plastic boxes and stored at the mentioned temperatures.
Ten carrots were used per treatment. Additionally, 10 carrots were wounded as described but not inocu-
lated and stored at 20°C for control. The storage time was 0, 2 and 6 months (Tab. 1). The development
of mycelium was monitored by daily measurement of the lesion diameter with a digital calliper. Lesion
diameter was measured in the width (D1) and in the length (D2). The measurements were taken at least
20 days and at the most 60 days.
3. Results and discussion
3.1 Comparison of destructive (conventional) and non-invasive methods for characterizing apple fruit
quality
3.1.1 Firmness
A close correlation between destructive (penetrometer) and non-destructively measured rmness values
(IFD) in Golden Delicious fruit could be determined.
The highest correlation occurred at harvest (R2=0.99) when the fruits were still rm. -
Fruit storage at high temperature (20°C) caused a clear decline in rmness and lower R2 valu- - es.
Correlation was still high for low storage temperature (R2 = 0.85 for 10°C) as well as for high - temperature (R2=0,87 at 20°C).
Innofruit & QualiTools
technical report
235
y = 5,6757x + 13,622
R
2
= 0,9932
51
52
53
54
55
56
57
58
59
60
61
55,5 66,5 77,5 88,5
firmness (destructiv)
firmness (non-destuctive)
Conclusion: The non-invasive technique with the “Intelligent Firmness Detector” is an appropriate me-
thod for evaluation of rmness in rm fruits at harvest as well as in softer fruits in the more advanced
maturity stage in the post-harvest period
3.1.2 Colour
Relationship between destructive and non-destructive colour measurement technologies under shelf life
conditions:
The senescence-induced breakdown - of chlorophyll content in the fruit skin
could be successfully monitored with
the non-destructive methods, e.g. laser
induced chlorophyll uorescence (Mi-
niVeg; F730) and light remission (PA;
NDVI).
The sun-lit and shaded sides of both - fruit cultivars differed in their chloro-
phyll content, and these differences
could be detected by both uorescence
and remission techniques around har-
vest time.
In cv. “Jonagold” the red anthocyanin - pigmentation could be directly detected
by NAI index and indirectly by higher
F730 values.
In summary, it could be documented that
in both tested apple cultivars difference in
ground colour due to chlorophyll decline du-
ring ripening could be measured with the
MiniVeg-System as well as with the Pigment
Analyser. Both technologies are suitable for
evaluation of green and red fruit colour as
attribute for product quality.
storage temperature 10°C
y = 3,7599x + 27, 803
R
2
= 0,8505
48
50
52
54
56
58
60
62
55,5 66,5 77,5 88,5
firm ness (de structiv)
firmness (non-destructive)
storage temperature 20°C
y = 6,6592x + 2,1851
R
2
= 0,8731
0
10
20
30
40
50
60
70
0 2 4 6 8 10
firmne ss (destructiv)
firmness (non-destructiv)
at harvest
Figure 2:
Correlation between rmness of apple fruits (“Golden
Delicious”) measured with a destructive penetrome-
ter and a non-invasive method (IFD) at harvest and
after storage at different temperatures
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6
0,8
1,0
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6
0,8
1,0
18 Oc t22 Oct 26 Oc t30 Oct 3 Nov 7 Nov 11 Nov15 Nov
0
1
2
3
4
5
6
7
18 Oc t22 Oct 26 Oc t30 Oct 3 Nov 7 Nov 11 Nov15 Nov
0
1
2
3
4
5
6
7
NDVI [rel. units]
sh ade si de
su nlit side
Gold en Del eciou s Jonagold
Chlorophyll content [µg cm
-2]
0
200
400
600
800
1000
1200
1400
1600
0
200
400
600
800
1000
1200
1400
1600
F730 [rel. units] at 5 kHz
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6
0,8
1,0
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6
0,8
1,0
18 Oc t22 Oct 26 Oc t30 Oct 3 Nov 7 Nov 11 Nov15 Nov
0
1
2
3
4
5
6
7
18 Oc t22 Oct 26 Oc t30 Oct 3 Nov 7 Nov 11 Nov15 Nov
0
1
2
3
4
5
6
7
NDVI [rel. units]
sh ade si de
su nlit side
Gold en Del eciou s Jonagold
Chlorophyll content [µg cm
-2]
0
200
400
600
800
1000
1200
1400
1600
0
200
400
600
800
1000
1200
1400
1600
F730 [rel. units] at 5 kHz
Figure 3: Chlorophyll decline during storage measured with
destructive and non-destructive methods
Innofruit & QualiTools
technical report
236
3.1.3 Firmness and Colour
All measured destructive and non-destructive values showed a good correlation between fruit colour as
external quality attribute and rmness as internal attribute. A close linear regression between the de-
structively measured chlorophyll content in the fruit skin and the rmness could be established both in
“Golden Delicious” and “Jonagold” (Figure 4). The correlation of data was closer when specically refer-
ring to the sun-exposed or shade side (Table 1).
Table 1: Correlation coefcient (r) between chlorophyll
content and rmness in the fruit skin of two apple
cultivars
Firmness
cv. / side sun side
n=80 shadow side
n=80 both
n=160
Golden D. 0,78 0,77 0,77
Jonagold 0,74 0,76 0,65
Also, a very close correlation was found between rm-
ness and the non-invasively estimated fruit colour as
measured with the Pigment Analyser (NDVI) and the
MiniVeg (F740, Table 2).
Table 2: Correlation coefcients (r) between fruit rmness and fruit colour (NDVI and F740)
Golden Delicious Jonagold
Cultivar/Side Sun (n=80) Shadow (n=80) both (n=160) Sun (n=80) Shadow (n=80) both (n=160)
NDVI 0,81 0,73 0,76 0,72 0,78 0,70
F730 (5kHz) 0,72 0,69 0,69 0,65 0,68 0,61
Compared with conventional practices all investigated non-invasive methods are appropriate techniques
for evaluating important fruit quality attributes. The collected data may be used to validate the existing
model for predicting product quality at any point of the supply chain and to forecast shef life period.
3.2 Carrot black rot disease progress in time at different temperatures
Temperatures below 2°C completely inhibit the lesion-development by
C. elegans
and
C. thielavioides
although the carrots were inoculated with densely concentrated suspensions conidia of the pathogens.
A temperature of 4°C can suppress the development of
C. thielavioides
for more then a month. The
same temperature can completely suppress the development of
C. elegans
for more then two months.
At a temperature of 8°C, after 10 days at earliest, the development of
C. elegans
and
C. thielavioides
occurred independent of the duration of storage. The rst symptoms (conidia) of
C. elegans
and
C.
thielavioides
incubated at 20°C were visible after 4 days (Fig. 5). Incubation at 20°C for one month
allowed lesion development to 6-8 mm diameter.
Independent of the site of inoculation (top, medium, peak) the lesions of
C. thielavioides
showed no
signicant differences in size by the development. In contrast the development of
C. elegans
at 20°C
was faster on the top compared to the center (medium) and the root tip (peak).
Figure 4: Relationship between chlorophyll
content and rmness in the fruit skin of
two apple cultivars
Innofruit & QualiTools
technical report
237
Figure 5: Development of
C. thielavioides
by different temperatures
At temperature below 8°C, there was no signicant difference in lesion development between
C. elegans
and
C. thielavioides
. At 20°C the development in width increased faster than in length. This trend has
been observed in all experiments.
The presented results shows that cooling of carrot throughout the whole chain, harvest, storage, sorting,
producing, POS and at customers home is essential for maintaining their quality. A temperature below
8°C suppresses the development of
C. elegans
and
C. thielavioides
on carrots during at least ten days.
If the customer leaves the carrots in the fridge the appearance of visible symptoms of either
C. elegans
or
C. thielavioides
should not occur within one week.
4. Summary and Conclusion
Availability of suitable non-destructive techniques for determining quality attributes is a prerequisite for
development and validation of models as basis for product quality prediction at any point of the supply
chain. In this project the non-invasively determined parameters of apples and tomatoes showed a close
correlation with data obtained with conventional procedures; therefore they can be recommended for
application in practice.
The Intelligent Firmness Detector is suitable for rmness measurements of spherical fruits like apple and
tomato with a minimum diameter of 4 cm. However, the Greefa instrument is not applicable to fruits
with an irregular shape like pears, avocados or cucumbers or to small fruit like plums and radish. Other
restrictions are low-rmness fruits like berries (strawberries or currants) due to impinging force of the
measurement device. In this case, a non-invasive instrument is the method of choice. Despite of the
representative and reproducible results in comparison with conventional rmness measurement devices,
the IFD is not an all-purpose or universal instrument which could be used by producers, sorters/packers,
distributors or retailers for a wide range of goods. Moreover, the IFD at the current developmental stage
is not a portable or handheld instrument because of its size, weight and sensitivity to vibrations. In the
present form, the IFD is well-suited for research purposes or as integral unit of sorting machines. It has
the potential for supplementing other sensors or technologies for automatic evaluation of external and
internal product quality characteristics.
For fruit colour estimation the pigment analyzer and MiniVeg proved applicability because of close corre-
lations of data with those of conventional methods. Both the PA and the MiniVeg are prototypes at the
Innofruit & QualiTools
technical report
238
advanced stage with lots of potential for precise determination of product quality in fruits and vegetab-
les. Furthermore, LIF offers the chance to detect
physiological diseases, e.g. internal browning, water core, sunburn and others - pathogen infections pre- and post-harvest long before damages or disorders becomed appa- - rent. Therefore this method is of further interest to horticultural science and practice.
More research driven by the project partners is on the way. This will open-up potential and perspectives
for introducing and establishing product quality models in horticultural supply chains. Furthermore, a
quantitative understanding of the quality change in time is another prerequisite for the development and
validation of models as a tool for quality management.
In this project the black rot disease progress (caused by
Chalara sp
.) in time as affected by temperature
was documented for the rst time. This quantitative knowledge can be used for practical carrot quality
postharvest management as well as for respective modelling purposes.
Modelling product quality
From the present work, the potential of process-oriented modelling becomes evident. The specic
knowledge of experts and specialists, whether scientic (theory), practical (empirical) or commercial
(application) can be used and is being applied to develop models on product quality and behaviour
that span the complete range of transregional supply chains. It is possible to include effects of seaso-
nal (within and between), regional and management practices. Provided that mechanism upon which
the models are based, do reect (more or less) the process occurring in the produce, the parameters
estimated are valid over the seasons and regions of provenience. That really offers the opportunity for
modelling and optimising supply chains. Of course a lot of work still needs to be done to achieve that
goal of modelling: predict future behaviour in any circumstance, from any region, grown in any season.
More research is necessary, more experiments need to be conducted, but from these few examples
we can deduce the framework for this new approach. And it has to be quite different compared to the
traditional research setups applied up till now. And the basis for this fascinating development has been
established by this project activity.
Innofruit & QualiTools
technical report
239
References
Blanke, M.M. (1992). Determination of chlorophyll using DMSO. Wein-Wissenschaft 47, 32-35.
Blanke, M.M. (2006). Nicht-invasive Zucker- und Festigkeitsmessung von Früchten. Erwerbs-Obstbau, 48, 31-37.
Courcoux, P., Chaunier, L., Lourdin, D., Della Valle, D., Nicod, H. & Semenou, M. (2005). Crispness of cereal products: Correlati-
on with descriptive analysis, sensory and acoustic measurements. Acta Hort. (ISHS) 674:521-522.
Early, R. (2005). Good agricultural practice and HACCP in fruit and vegetable cultivation. (In: W. Jongen (Ed.). Improving the
safety of fresh fruit and vegetables (pp 229-266), Cambridge, England, Woodhead Publishing Ltd.
Heller, W., Bosshard, E. & Barofo, C. (2005). Zur Situation der Schwarzeckenpilze in der Schweiz im Jahr 2005. Der Gemüse-
bau 3/2005, 5-6.
Höhn, E. & Winkler, S. (2003). Comparison of different types of rmness measurement. Proc. EUFRIN Workshop on Fruit Quali-
ty, June 2003, Bologna, 143-144.
Johnson, D.S. & Dover, D.J. (2005). Does ´acoustic rmness´ relate to sensory perception of apple texture? Acta Hort. (ISHS)
682:1395-1402.
Hertog, M.L.A.T.M. (2002). The impact of biological variation on postharvest population dynamics. Postharvest Biology and
Technology, 26, 253-263.
Hertog, M.L.A.T.M. (2006). Managing biological variation. In: A. C. Purvis et a. (Eds.), Acta Horticulturae 712: Proc. IVth IC on
MQUIC 2006.
Hertog, M.L.A.T.M., Lammertyn, J., Desmet, M., Scheerlinck, N. & Nicolaï, B.M. (2004). The impact of biological variation on
postharvest behaviour of tomato fruit. Postharvest Biology and Technology, 34, 271-284.
Kuckenberg, J., Tartachnyk, I. & Noga, G. (2008). Evaluation of uorescence and remission techniques for monitoring changes
in peel chlorophyll and internal fruit characteristics in sunlit and shaded sides of apple fruit during shelf life. Postharvest
Biology and Technology, 48, 231–241.
Lana, M.M., Tijskens, L.M.M. & van Kooten, O. (2005). Effects of storage temperature and fruit ripening on rmness of fresh cut
tomatoes. Postharvest Biology and Technology, 35, 87-95.
Schouten, R.E., Otma, E. C., van Kooten, O. et al. (1997). Keeping quality of cucumber fruits predicted by the biological age.
Postharvest Biology and Technology, 12, 175-181.
Schouten, R.E. & van Kooten, O. (1998). Keeping quality of cucumber batches: Is it predictable? Acta Hort. (ISHS) 476:
349-355.
Snoekx, P., De Baerdemaeker, J., Hertog, M. & Nicolaï, B. (2005). MODEL FOR THE PREDICTION OF HORTICULTURAL PRODUCT
QUALITY EVOLUTION ALONG DIFFERENT SUPPLY CHAIN SCENARIO PATHS. Acta Hort. (ISHS) 674: 359-366.
Streif, J. (1996). Optimum harvest date for different apple cultivars in the ‘Bodensee’ area. In: de Jager, A., Johnson, D. and
Hohn, E., (Eds), COST 94. The Postharvest Treatment of Fruit and Vegetables: Determination and Prediction of Optimum
Harvest date of Apple and Pears, ECSC-EC-EAEC, 1996, Brussels, 15-20.
Tijskens, L.M.M. & Konopacki, P. (2003). Biological variance in agricultural products: theoretical considerations. (In: J. Ooster-
haven & H. W. Peppelonbos (Eds.), Acta Horticulturae 600: VIII International Controlled Atmosphere Research Conference
CA2001 (pp 661-669), Rotterdam, The Netherlands, 8-13 July.,
Tijskens, L.M.M., Konopacki, P. & Simičič, M. (2003). Biological variance, burden or benet? Postharvest Biology and Technology
27, 15-25.
Tijskens, L.M.M., Heuveling, E., Schouten, R.E., Lana, M.M. & van Kooten, O. (2005). The biological shift factor: biological age
as a tool for modelling in pre- and postharvest horticulture. Acta Hort. 687: 39-46.
Tijskens, L.M.M., van Kooten, O. & Schouten, R.E. (2006). Modelling for Globalisation. Acta Hort. 712: 51-55.
Innofruit & QualiTools
View publication statsView publication stats
