From Farm to Fork: Traceability based on RFID
A Proposal for Complete Traceability in the Wine Sector
Iñigo Cuiñas, Isabel Expósito, José Antonio Gay-Fernández, Ana V. Alejos and Manuel G. Sánchez
Dept. Teoría do Sinal e Comunicacións, Universidade de Vigo, r. Maxwell, s/n 36310 Vigo, Spain
Keywords: RFID, Traceability, Tag, Wine.
Abstract: This paper highlights the objectives and activities of the European project “RFID from Farm to Fork”,
which is focused on the food industry traceability. The final goal is to extend the traceability information to
the final consumer, so these persons could feel the confidence on the food origin like our grandparents did
when they lived in a less globalised World. The activities of the project in a winery at Ribeiro denomination
of origin (Spain) at both vineyards (by means of wireless sensor networks) and the wine production process
(by RFID technology) are also presented along the paper, in order to show some of the project outcomes.
1 INTRODUCTION
European Regulation CE178 (Regulation EC
178/2002, 2002) stated that the capacity of tracing
and tracking a food along every stages of
production, transformation, and distribution is a key
question in the food sector. And it is important both
in terms of quality and safety, and in terms of
technology.
Thinking on quality and safety, the objective is
to recover the confidence of the consumers in the
products they purchase. That confidence on the food
products has evolved during last times, as the way
we acquire the food has changed a lot as our
societies advance and move their activity and
industrial sectors balance. Some years ago (and in
several countries not so many years ago), consumers
were very confidents on the origin of the food they
eat: people bought the food directly to the producer,
who belonged to a family known for generations; or
the number of intermediaries was very reduced, and
they were met by the buyers or neighbours, as well
as the distance covered by the product from its
origin (Cuiñas et al, 2011b). That time, people
bough wine bottles (or perhaps barrels) directly to
the winery, which was very confident. Perhaps the
buyers had previously sawn the farmers taking care
of the vineyards, or carrying the grapes, or washing
the barrels and bottles. Probably, the parents and
grandparents of these consumers had been clients of
the parents or grandparents of the winery holders!
Or, in the case they bought the wine in a shop, they
most likely met the seller for long ago, and also the
seller met the wine producer. And the same occurred
with meat, cheese, fish, and all the food they
consumed.
Besides, the transportation steps in the
production chain were short, or even there were no
transportation, as the local products dominated the
shops and markets. Possible problems related to
transport and delivery conditions (temperature,
conservation, etc.) had no space in the consumers
mind some years ago.
Currently, this situation has radically changed, as
our shopping habits are not the same of some years
ago. Nowadays, the small confident shop, or the
traditional market, is mainly reserved for special
days, for people looking for delicatessen groceries,
or for elderly that feel better maintaining their
traditional lifestyle. Most of the people generally
buy their food products in super or hypermarkets,
where the i.e. cheese is presented in clean and
aseptic plastic boxes, and labelled providing some
information. But we have lost the contact to the
farmer that took care of the cow or the sheep! So, we
receive a lot of information, but we do not have
confidence in the origin and in the process suffered
by the piece of cheese or meet, or the wine bottle,
we are buying. Something similar occurs with other
products.
Within this context of confidence loss, the price
of the product could be a determinant factor to
decide the acquisition of a product among the
265
Cuiñas I., Expósito I., Gay-Fernández J., V. Alejos A. and G. Sánchez M..
From Farm to Fork: Traceability based on RFID - A Proposal for Complete Traceability in the Wine Sector.
DOI: 10.5220/0004022302650270
In Proceedings of the International Conference on Signal Processing and Multimedia Applications and Wireless Information Networks and Systems
(WINSYS-2012), pages 265-270
ISBN: 978-989-8565-25-9
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
variety the supermarket offers us. This situation
gives an advantage to goods produced by large
companies, which could assure the quality along the
production chain, but also to those produced in
countries with lower salary costs or with less strict
sanitary laws, where they could manufacture at
lower costs.
Knowing the situation related to control the
quality and safety of the goods we buy for eating,
the high quality products developed by small
companies seem to be in disadvantages against these
products with less quality but also less price. And
this situation is becoming worse in an economic
crisis context. The chance for small and medium
enterprises (SME) in developed countries is the
incorporation of new technologies to focus their
activities to the production of added value products:
in terms of quality, in terms of origin guarantee, in
terms of deep information, or combining all of them.
An SME could not compete against non developed
country companies in price, but they could in quality
and origin control. Besides, these SME are in
disadvantage against large companies in terms of
assuring complete traceability.
The proposal of the European project “RFID
from Farm to Fork” (RFID-F2F) is focused on
SMEs and on a production sector: the food industry
in Europe. We have important salary costs, which
could be a problem to fight against producers from
other countries, but we also have strong experience
in origin guarantee systems as well as on sanitary
control of the food, which is assumed to be an
advantage in demonstrating the quality of the
product. The goal is to give this information to the
consumers… and, then, the consumers could select
this higher quality product! Large companies could
assure the traceability from the farm to the
consumer, as they control all the stages of the
business chain: farming, processing, transportation
and delivering. SME are only present in one or two
of these elements of the chain, so they need an
external umbrella to hold the traceability links
among the different agents.
2 THE “RFID-F2F” PROJECT
The project “RFID from Farm to Fork” is a CIP-
Pilot action involved within the 7th Frame Work of
the European Commission. The consortium is led by
the University of Wolverhampton (United
Kingdom), being the other members universities
from Spain (Vigo and Politécnica de Cartagena),
Slovenia (Ljubljana) and Italy (Salento), companies
(Treviso Tecnologia, Santer Reply, IDxS BVBA), or
institutions (EuroFIR, Institute of Food Research).
So, the represented European countries are five:
United Kingdom, Spain, Italy, Slovenia and
Belgium.
The action itself looks for the extension of radio
frequency identification (RFID) technologies
(Europe’s Information Society web page, 2011)
along the complete food supply chain: from the
farms where cows, fishes, sheep, grapes, etc. grow;
to the final consumer at supermarkets, including all
the intermediate stages: transport, manufacturing
processes, storage. The main objective is the use of
only one technology to perform a complete
traceability, recording data at each stage.
The final consumers could obtain different data
about the whole process experienced by the product
they are buying, just by moving the object (labelled
with a RFID tag) in the vicinity of a RFID reader,
which can be installed in the supermarket. An
alternative could be products provided by QR codes
or NFC tags that could be read from an application
at each personal smart phone.
The individual identification of a product allows
the software to obtain a complete traceability report
from a central database, and to bring the consumer
this information. Thus, the buyer could know what
happens with i.e. the wine at each instant of its
production, and also the conservation and
transportation conditions from the winery to the
shop, although each production stage would be
conducted by different agents.
Besides, each of the producers along the supply
chain could use the identification by radio frequency
to control its production and storage, and to know
some previous information of its ingredient matters.
The project involves both the design of the
complete system and its tests at different stages of
the chain: fishing companies, wine producers, food
transporters, and final users, in order to define the
actual interest of the system, its performance, and its
advantages and disadvantages (Swedberg, 2011).
After the tests or pilot experience, the consortium
will have valued information on the possibilities of
implementing RFID technologies for tracing and
tracking food products, as well as a deep evaluation
from the companies. This information would be
transferred to the European companies, and
indirectly to the whole society as a return of
investment of the project efforts.
The project will show the ability of RFID
technologies to make a return on investment for
SMEs in the food industry, as well as to provide
large information to the consumers (RFID-F2F web
WINSYS2012-InternationalConferenceonWirelessInformationNetworksandSystems
266
page, 2011). The opportunities for such a return on
investment arise from the increment of productivity
due to authentication, quality control, wastage
reduction, and energy optimization. Until now, these
advantages have been demonstrated in large
organizations, which have control over most or all of
the value chain and are in a position to make an end-
to-end investment. Vice versa, the “farm to fork”
traceability system has not yet been adopted by
independent SMEs, which only participate in one
stage of the value chain. By linking RFID and sensor
network technologies with an Europe wide database
as EuroFIR (EuroFIR web page), which can store
the exact history of any food product, SMEs will be
given the opportunity to optimize their own business
process to maximize return. In addition, a pan-union
resource will be created allowing producers to
demonstrate the quality and freshness of their
product, which will have the effect of increasing
consumer confidence as well as producer margins
(Cuiñas et al., 2011a)
It is known that the use of RFID in the food
chain has been previously tested, and it could
obviously provide a competitive advantage to the
involved companies (Gandino et al., 2007) (Pérez-
Aloe et al, 2007) (Fenu and Garau, 2009). However,
the radio propagation in presence of liquids presents
important differences compared to other more
friendly environments (Dobkin and Weigard, 2005).
Wine bottles, as an example, suffer from this
disadvantage, which needs to be evaluated
previously to incorporate this technology to the wine
industry.
3 THE RIBEIRO WINE PILOT
Denomination of origin is the name used in Spain to
define appellation regions: areas where a specific
product is controlled, taking care of the geographical
origin of the matters, the variety of the plants or
animals involved, the processes employed during the
elaboration of the food product, and also aspects as
bottling, labelling, storing and delivering. Among
various Spanish wine denominations of origin,
“Ribeiro” is a controlled wine producing region
located in Galicia, the North Western area of Spain.
In a pilot company, both WSN and RFID systems
have been deployed at the vineyards and winery,
respectively.
The selected environment to deploy the WSN is
a vineyard located in a mountain side from
Ribadavia. This vineyard is property of the winery
company.
The grapes involved in the production of Ribeiro
wines have to come from a strictly delimited area.
So, a requirement is to assure the location of the
vineyards by means of precise methods. The control
of different weather parameters is a must. The region
is very rainy in winter, but it is also very warm in
summer. The high humidity levels, in co-ordination
with the sunny time, could lead to problems with
plagues which are the main trouble during the
growing of the grapes. The installation of a wireless
sensor network (WSN) with adequate sensors help
the farmers to develop tailor made strategies to keep
the health of their plants.
During the wine manufacturing, at the winery, a
fast and comfortable way to control all the
movements of liquids among the different barrels,
across filters, cleaners, and other machines is one of
the claims of the winery managers. The proposal of
a RFID based traceability system, operated by a
handheld RFID reader appears to be a good solution
for this task. Water resistant RFID tags will be glued
to each machine and each barrel, and a three steps
standard event will be defined to describe the
movement of liquids from tank A (first step) to tank
B (second step) across the machine M (third step).
Some chemical data would be added manually to the
database at different stages along the manufacturing
of the wine.
The management of the bottle stock by
individually tagging the bottles appear to be more
complicated: there were some reading problems
around the bottles (Expósito and Cuiñas, 2011) and,
which is no less important, the cost of each
individual tag is still too large to be economical
assumable by the winery.
4 WIRELESS SENSOR
NETWORK AT THE
VINEYARDS
4.1 Network Description
A system following the ZigBee standard has been
installed during November 2010 last week, and it
has been working continuously since that date. A
coordinator node and six regular nodes compose the
communication facilities of the WSN. Different
sensors are connected to the regular nodes: soil
moisture and temperature, soil water content,
ambient temperature and humidity, leaf wetness,
solar radiation, and one weather station. The nodes
assure the communication of data among them and
FromFarmtoFork:TraceabilitybasedonRFID-AProposalforCompleteTraceabilityintheWineSector
267
towards the coordinator node. Each node collects
regularly the data from the sensors it has plugged in,
and it also transmits these data towards the
coordinator node. This coordinator connects to a
gateway to extract the information from the WSN to
a database installed at the winery or a management
place. The deployment and performance of such a
network has been widely described in (Gay-
Fernández and Cuiñas, 2011).
Prior to the deployment, large tests were
performed in controlled environments: both
propagation measurement campaigns and laboratory
experiments were carried out. The propagation
campaigns were focused in defining attenuation
models adapted for different environments (Gay-
Fernández and Cuiñas, 2011) (Gay-Fernández et al.,
2010). These studies allowed us to decide the
distances between nodes, which resulted to be up to
250 meters, depending on the environmental
conditions. The laboratory tests had the objective of
measuring the node energy consumption. We
observed that the nodes could work up to 4 months
without solar energy. As the individual solar panels
recharge the batteries during sunny (or even clear)
days, the node life would not be conditioned by the
energy supply.
4.2 Connectivity Aspects
The architecture of the global system is depicted at
Figure 1, involving the own WSN and the link to
systems placed out of the vineyards, where the data
has to be stored and then analyzed. The connectivity
at both stages has to be considered.
Figure 1: Global system architecture.
The connection among nodes is driven by the
ZigBee protocol, based on a mesh topology: if one
node fails, the network is automatically
reconfigured, looking for new paths to reach the
isolated nodes. During the operation of the WSN we
detected some connectivity fails due to the growing
foliage of the vineyard, but also due to the presence
of trees cutting the line of sight of the radio links.
The connection from the coordinator node to the
database could be more complicated, as external
power supply could be needed. The adopted solution
was the use of the GPRS network to extract the data
from the coordinator to the database via the Internet,
running tailor-made software on a local PC.
5 RFID DEPLOYMENT AT THE
WINERY
During the elaboration of the wine from the
harvested grapes, a lot of processes have to be
applied within the winery. The objective of the
RFID deployment at the winery is to register the
different traceability events suffered by the product,
in order to know what, where and when each event
occurs.
5.1 Hardware Definition
The hardware definition includes the selection of the
RFID tags and the RFID readers most adequate for
being employed along the wine production chain,
within the winery. The RFID activities begin at the
vineyard, where the use of a hand held reader to
read/write the tags at containers or trucks after the
harvesting appears to be better option than fixer
interrogator. Water resistant encapsulated tags
would be needed at this and other stages, in order to
survive the ambient conditions.
Figure 2: Tests on wine tanks.
At the winery, both hand held and fixed readers
will be needed: the hand held reader to read the tags
on grape containers, wine tanks and other
equipment/machines along the wine processing; and
the fixed readers will be mounted on conveyor belts,
forklifts and arcs at storage doors.
Previously to the selection of hardware and tags,
some tests have been performed, using tags glued on
WINSYS2012-InternationalConferenceonWirelessInformationNetworksandSystems
268
wine m
e
the test
o
tag ori
e
reading
differen
t
Figure
2
maximu
m
tag glue
d
in Tabl
e
suffered
Table 1:
R
the tag.
Orien
t
0º/90º
n
0º on
m
30º on
60º on
90º on
120º o
n
150º o
n
The
anechoi
c
readabil
i
b
ottles,
with up
measure
m
tailor-m
a
measure
m
read in (
E
Thes
p
ower a
f
wine is
l
empty b
o
b
ottle is
arc widt
h
5.2
S
The bu
s
definitio
RFID a
p
attack s
u
and the
n
catalog
u
p
roducti
o
consists
the tan
k
equipm
e
standar
d
Mos
t
supply
c
handhel
d
e
tallic tanks a
n
o
n metallic t
a
e
ntation, as
w
distances. W
e
t
encapsulat
e
2
depicts an
i
m
reading d
i
d
vertically o
n
e
1. This is
p
by the tag w
h
R
eading range
s
t
ation related to
t
n
on-metal surfac
m
etal surface
metal surface
metal surface
metal surface
n
metal surface
n
metal surface
second test
c
chamber. I
t
i
ty pattern ar
o
and two em
p
to seven R
F
m
ent system
a
de MATLA
B
m
ents aroun
d
E
xpósito and
e previous t
e
fter reading
a
l
ess than 25
%
o
ttle. Besides
,
modified in i
h
appears to
b
S
oftware
s
iness proces
s
n, which all
o
p
plication for
u
ch task was
t
n
to adapt a
l
u
e of standar
d
o
n chain fo
l
on moving a
k
/barrel A to
e
nt as filter
s
d
event could
b
t
of the dat
a
c
hain are goi
n
d
reade
r
, as it
n
d on bottles
.
a
nks was to
d
w
ell as to
d
e
used a han
d
e
d tags for
i
nstant of su
c
i
stance was
o
n
the tank, as
p
robably due
h
en it is horiz
o
s
as a function
o
t
he vertical
e
has been
m
t
consisted
o
o
und differen
t
p
ty used as
r
F
ID tag mod
e
was control
l
B
scripts. A
d
d
white wine
Cuiñas, 2011
e
sts indicates
a
RFID tag
o
%
of the same
,
the reading
p
ts shape, as
w
b
e reduced.
s
was the b
a
o
ws the pro
g
tracking sys
t
t
o define so
m
l
l the event
d
s. Many pro
c
l
low a sim
p
volume of wi
n
the B passi
n
s
, chutes, d
e
b
e easily defi
n
a
generated
a
n
g to be col
l
is more com
f
.
The objecti
v
d
etermine the
d
efine maxi
m
d
held reader
this experi
m
c
h campaign.
o
btained wit
h
could be che
c
to the curv
a
o
ntally locate
d
o
f the orientati
o
Reading range
162 cm / 160 cm
81 cm
78 cm
73 cm
53 cm
53 cm
73 cm
m
ade withi
n
o
n measuring
t
bottles (fou
r
r
eference) ta
g
e
ls. The com
p
l
ed by mean
d
escription o
f
bottles coul
d
)
.
that the rece
o
n a bottle fu
l
magnitude o
n
p
attern aroun
d
w
ell as the rea
d
a
sis for an e
g
ramming o
f
t
ems. The w
a
m
e standard e
v
d
efinition to
c
esses along
w
le scheme:
n
e (or must)
f
n
g through s
e
canters,... S
o
n
ed.
a
long wine
p
l
ected by usi
n
f
ortable to be
u
v
e of
best
m
um
and
m
ent.
The
h
the
c
ked
a
ture
d
.
o
n of
n
an
the
r
full
g
ged
p
lete
n
s of
f
the
d
be
e
ived
ll of
n
an
d
the
ding
e
vent
f
the
a
y to
v
ents
this
w
ine
they
f
rom
s
ome
o
, a
p
ilot
n
g a
u
sed
alo
n
ha
n
sep
p
ro
the
F
an
d
an
col
l
ma
n
ch
e
sel
e
cur
r
ob
s
htt
p
(th
e
Th
e
det
e
Th
e
p
h
o
thi
s
alo
n
inf
o
we
b
inf
o
of
t
n
g the winer
y
n
d held appli
c
arate forms
duction proc
e
initial displa
y
F
igure 3: Displ
a
After the rea
d
d
intermediat
e
EPCIS co
m
l
ected data.
n
agement op
e
mical analysi
Final consu
m
e
cting goods
r
ent grocerie
s
s
erved at th
e
p
://www.yout
u
Some produ
c
e
RFID versi
o
e
consumers
e
ct the NFC
s
e
informatio
n
o
ne to the RF
I
s
portal is us
e
n
g the bus
o
rmation, wh
i
b
page. Thu
s
o
rmation on t
h
t
he wine bottl
e
y
, at different
c
ation has b
e
depending
e
ss to be perf
o
y
of such appl
i
a
y of the hand
h
d
ing of origi
n
e
action, the
R
m
pliant XM
L
Extra info
r
e
ration coul
d
s
data).
m
ers would
h
among a la
r
shops or su
p
e
demonstrat
i
u
be.com/watc
c
ts could be l
a
o
n for low fre
q
would use
t
s
ignals or to
i
n
at such ta
g
I
D-F2F web s
e
e
d to ask the
d
i
ness chain
i
ch is elabora
t
, final cons
u
h
e different s
t
e
they have o
n
operation ar
e
e
en program
m
on the ste
p
o
rmed. Figur
e
ication.
h
eld reader app
l
n
and destinat
i
R
FID reader
L
file conta
i
rmation for
d
also be a
d
h
ave the pos
s
r
ge catalogu
e
p
ermarkets, a
s
i
ve project
v
c
h?v=lmvHo
h
a
belled with
N
q
uencies) or
Q
t
heir smart
p
interpret the
Q
g
s would co
n
erver. The co
d
database at e
a
for its tr
a
a
ted and pres
e
u
mers obtain
t
ages of the p
r
n
their hands.
e
as. So, a
m
ed with
p
of the
e
3 shows
l
ication.
i
on tanks,
generates
i
ning the
internal
d
ded (e.g.
s
ibility of
e
at their
s
could be
v
ideo a
t
:
m
3SYA.
N
FC tags
Q
R codes.
p
hones to
Q
R code.
n
nect the
d
e sent to
a
ch agent
a
ceability
e
nted as a
complete
r
oduction
FromFarmtoFork:TraceabilitybasedonRFID-AProposalforCompleteTraceabilityintheWineSector
269
6 CONCLUSIONS
The European project “RFID from Farm to Fork”
activities and objectives have been presented along
this paper, as an application of RFID technologies to
provide traceability along the complete food chain,
from producers to final consumers.
A pilot experience developed in a winery at
Ribeiro area, in Spain, has been also presented, in
order to show a specific installation. Previously to
this development, various tests have been carried out
in both WSN and RFID technologies, and their
results have been also commented.
Related to WSN deployment, propagation
models were defined from the results of large
measurement campaigns. They were useful to define
the distances between nodes in the network, and to
install the complete system.
The readability of RFID tags on metallic tanks
and on full wine bottles has also analyzed, and the
results were valid to determine the better allocation
of tags on such elements. The read distance when
the tag is glued on a metallic tank is reduced in more
than 50% compared to non metallic. And this
reduction also depends on the orientation the tag is
installed.
Related to the readability patterns around wine
bottles, the effect of the wine is very significant in
terms of received power and in terms of the shape of
the pattern. When the bottle is empty, the RFID tag
could be read all around, at enough power levels.
When the bottle is full of wine, the received power
from the tag is reduced significantly, whereas there
are various directions at which it is not possible to
read the tag: the shape of the readability patterns
change.
ACKNOWLEDGEMENTS
This work has been supported by the European
Commission (CIP-Pilot Actions), under the project
“RFID from Farm to Fork”, grant agreement number
250444. The authors would like to thank the winery
“Vitivinícola do Ribeiro” for their enthusiastic help
along the pilot project.
REFERENCES
Cuiñas, I., Catarinucci, L., Trebar, M., “RFID from Farm
to Fork: traceability along the complete food chain”,
Progress In Electromagnetic Research Symposium,
PIERS 2011, Marrakesh (Morocco), 2011.
Cuiñas, I., Expósito, I., Gay-Fernández, J. A., “The “RFID
from Farm to Fork” project proposal for food industry
traceability. A wine pilot example”, 5as Jornadas
Cientícias sobre RFID, Tarragona (Spain), 2011.
Dobkin, D. M., Weigand, S. M., “Environmental effects
on RFID tag antennas” IEEE MTT-S International
Microwave Symposium Digest, 2005.
EuroFIR (European Food Information Resource) web
page, http://www.eurofir.net
Europe’s Information Society web page, Thematic Portal,
http://ec.europa.eu/information_society/apps/projects/f
actsheet/index.cfm?project_ref=250444.
Expósito, I., Cuiñas, I., “RFID tag readability around
white wine bottles”, 19th International Conference on
Software Telecommunications and Computer
Networks, SoftCOM 2011, Split (Croatia), 2011.
Fenu, G., Garau, P., “RFID- based supply chain
traceability system”, 35th Annual Conference of IEEE
Industrial Electronics, IECON '09, pp. 2672-2677,
2009.
Gandino, F., Montrucchio, B., Rebaudengo, M., Sanchez,
E.R., “Analysis of an RFID-based Information System
for Tracking and Tracing in an Agri-Food chain”, 1st
Annual RFID Eurasia, 2007.
Gay-Fernandez, J. A., Garcia Sanchez, M., Cuiñas, I.,
Alejos, A. V., Sánchez, J. G. and Miranda-Sierra, J.
L., 2010, “Propagation Analysis and Deployment of a
Wireless Sensor Network in a Forest”, Progress In
Electromagnetics Research, PIER 106, pp. 121-145.
Gay-Fernández, J. A., Cuiñas, I., “Deployment of a
wireless sensor network in a vineyard”, International
Conference on Wireless Information Networks and
Systems, WINSYS 2011, Seville (Spain), 2011.
Perez-Aloe, R., Valverde, J. M., Lara, A., Carrillo, J. M.,
Roa, I., Gonzalez, J., “Application of RFID tags for
the overall traceability of products in cheese
industries”, 1st Annual RFID Eurasia 2007.
Regulation (EC) No 178/2002 of the European Parliament
and the Council of 28 January 2002, “laying down the
general principles and requirements of food law,
establishing the European Food Safety Authority and
laying down procedures in matters of food safety”,
Official Journal of the European Communities, 1
February 2002.
RFID-F2F (RFID From Farm to Fork) project web page,
http://www.rfid-f2f.eu/
Swedberg, C., “Ambitious European Project Traces Food
from Farm to Fork”, RFID Journal, article 8913,
October, 28
th
2011.
WINSYS2012-InternationalConferenceonWirelessInformationNetworksandSystems
270
