01 trazabilidad en alimentos

8/12/2019 01 Trazabilidad en Alimentos

1/6

Testing improvements in the chocolate traceability system: Impact on productrecalls and production efciency

Rolando Saltini, Renzo Akkerman*

Department of Management Engineering, Technical University of Denmark, Produktionstorvet 424, 2800 Kgs. Lyngby (Copenhagen), Denmark

a r t i c l e i n f o

Article history:

Received 26 February 2011Received in revised form11 July 2011Accepted 16 July 2011

Keywords:

TraceabilitySupply chainRecallSimulationChocolateFood safety

a b s t r a c t

The primary aim of food traceability is to increase food safety, but traceability systems can also bringother benets to production systems and supply chains. In the literature these benets are extensivelydiscussed, but studies that quantify them are scarce. In this paper we propose two hypotheticalimprovements of the traceability system within the chocolate production system and supply chain andwe illustrate the resulting benets by using a case study. Based on the case study, we quantify theinuence of these improvements on production efciency and recall size in case of a safety crisis bydeveloping a simulation tool. These results are aimed to illustrate and quantify the additional benets oftraceability information, and could help food industries in deciding whether and how to improve theirtraceability systems.

2011 Elsevier Ltd. All rights reserved.

1. Introduction

Dueto recent food scares, much attention has been given to foodsafety issues (e.g.Trautman, Goddard, & Nilsson, 2008). Food safetycrises like the bovine spongiform encephalitis (BSE) in the 1990sglobally caused economical losses for US$5.6 billion, and approxi-mately seven million people are affected by food borne illness eachyear in Europe (Kimball & Taneda, 2004; Sarig, 2003; World HealthOrganization, 2002, p. 1). The consequences of food safety crisescould be reduced with appropriate traceability systems that keeptrack of food trades and processing information along the foodsupply chain. With such traceability systems, the identication ofcauses could be faster, market withdrawals and/or recalls of theinvolved products could be reduced in size, and total costs of thecrises could be decreased.

1.1. Traceability in the food supply chain

In relation to food products, risk is dened by the EuropeanCommission as a function of the probability of an adverse healtheffect and the severity of that effect, consequential to a hazard(European Commission, 2002). Most food products containa certain degree of risk. The magnitude of risk directly affects safety

of the foods to public health and possibility of a food crisis that mayrequire a product recall. The costs of a product recall can varydepending on food category, magnitude of the recall and serious-ness of the food safety problem. A large share of the costs are oftencaused by indirect consequences, such as damage to the reputationof the brand and loss of market value, which can even lead to thebankruptcy of the brand (Onyango et al., 2008; Wilcock, Ball, &Fajumo, 2011). However, traceability systems themselves do notreduce the probability that food crises occur, they only reduce theirconsequences. The better and more accurate a traceability systemis, the faster a rm can identify a food safety problem, resolve it bywithdrawing or recalling the implicated products and effectivelyinvestigate what caused the problem (Karlsen, Donnelly, & Olsen,2011; Regattieri, Gamberi, & Manzini, 2007; Ruiz-Garcia, Stein-berger, & Rothmund, 2010).

In the food industry, raw material batches from differentsuppliers may be mixed. If a food safety problem comes froma certain raw material batch, all nished products containing rawmaterials from that batch have to be identied and recalled. Thus,the magnitude of a recall directly depends on batch dispersion inproduction and distribution. Recently, researchers found that thebest solution to reduce batch dispersion is to reduce processingbatch size and batch mixing. However, it was also found thatreducing batch size leads to losses in production efciency, due toincreased production setup times, setup costs, cleaning efforts, etc.(Dupuy, Botta-Genoulaz, & Guinet, 2005; Wang, Li, & OBrien, 2009;Rong & Grunow, 2010).

* Corresponding author. Tel.: 45 45254736; fax: 45 45256005.E-mail address:[email protected](R. Akkerman).

Contents lists available atScienceDirect

Food Control

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m/ l o c a t e / f o o d c o n t

0956-7135/$ e see front matter 2011 Elsevier Ltd. All rights reserved.

doi:10.1016/j.foodcont.2011.07.015

Food Control 23 (2012) 221e226
mailto:[email protected]://www.sciencedirect.com/science/journal/09567135http://www.elsevier.com/locate/foodconthttp://dx.doi.org/10.1016/j.foodcont.2011.07.015http://dx.doi.org/10.1016/j.foodcont.2011.07.015http://dx.doi.org/10.1016/j.foodcont.2011.07.015http://dx.doi.org/10.1016/j.foodcont.2011.07.015http://dx.doi.org/10.1016/j.foodcont.2011.07.015http://dx.doi.org/10.1016/j.foodcont.2011.07.015http://www.elsevier.com/locate/foodconthttp://www.sciencedirect.com/science/journal/09567135mailto:[email protected]


2/6

1.2. Value of traceability information

According to Golan et al. (2004) traceability consist of datacollection regarding the origin and processes of a product. Thus,having an accurate traceability system also means having anextended amount of data on the products. That is, all actorsinvolved in the supplychain of a food product collect and store datawhich, theoretically, could be available for all actors involved aswell as the consumers. Depending on the nature of the products,data collected by a single actor of a supply chain could be useful forother actors involved in the same supply chain, like food manu-facturersor traders. In modern food supplychains a huge amountofdata is collected but most of it is not exchanged between or notavailable for the different actors in the supply chain.

In the literature much material is available regarding the valueof traceability information for consumers, but few studies discussthis value for actors of the food supply chain which are not

consumers (Canavari, Centonze, Hingley, & Spadoni, 2010). Somerecent work did however focus on additional benets, like lowerinventory levels, faster detection of difculties in manufacturingprocesses by improved process control, decreased labor cost,improved operational planning, and increased efciency of logisticsand distribution operations (Alfaro & Rbade, 2009; Mai, Bogason,Arason, rnason, & Matthasson, 2010; Viaene & Verbeke, 1998;Wang, Li, & Li, 2009). To achieve these benets of traceability thekey issue is to integrate traceability with operations and supplychain management, and to increase the extent of traceabilityinformation being collected, which is currently often very myopicand only focused on tracing the origin of products. This lack incommunication might be due to the absence of an informationsystem able to handle data in an appropriate way, probably in turn

caused by the absence of an interest of the food companies in suchdata. Thus, the challenge here is to nd benets of having this dataavailable that could justify the investment needed in creatinga system which could increase the exchange of data between theactors in the supply chain, from food manufacturers to the nalconsumers. Furthermore, it is essential to be able to evaluate theexpected magnitude of such benets, and even though thesebenets of traceability are extensively discussed in the literature,only few papers actually quantity them (Akkerman, Farahani, &Grunow, 2010; Thakur, Wang, & Hurburgh, 2010; Wang, Li, &OBrien, 2009).

2. Objectives

The aim of this paper is to quantify and discuss the benets oftraceability improvements to food manufacturers. As the quanti-cation would not be possible on a completely theoretical level, wechose to focus on the supply chain of chocolate and the productionsystem of a chocolate manufacturer to be able to illustrate thetrade-offs and potential benets of improving traceability. Based onthis case study, we are able to illustrate two main aspects. First, wequantify the inuence of reducing batch dispersion on productionefciency and recall size. Secondly, we quantify the benets interms of reduction of recall size for possible improvements of thetraceability system used in the cocoa supply chain. Overall, we aimto demonstrate the additional value of traceability information inthe chocolate supply chain, as well as providing an inspirational

example case for other industries.

3. Materials and methods

3.1. Case study

The case study was constructedbased on available literature anda series of interviews with industry experts (from two differentchocolate manufacturers and a cocoa trading company). In order toprotect the condentiality of their contributions, data on theinvolvedrms is omitted. The entire supply chain of the case studyis shown inFig. 1and briey outlined below.

The supply chain of chocolate starts with cocoa farming. Cocoafarmers grow, harvest, ferment and dry the cocoa beans. Afterpacking the dried beans in 15-kg bags the cocoa farmers deliver thecocoa beans to local buying stations, which combine the bags ofseveral cocoa farmers and deliver them to a cocoa exporter (deMuijnck, 2005; Lain, 2001). Cocoa exporters combine the bags ofseveral local buying stations into batches which size typically varies

between 10,000 kg and 18,000 kg. The cocoa exporters organize theshipping to the chocolate manufacturer. It is assumed that thechocolate manufacturer produces monthly 741,000 kg of chocolate,packed in bars of 100 g. This quantity was chosen as it representsthe average of the monthly production of all Swiss chocolatemanufactures in 2008 (ChocoSuisse, 2009). The mass balance of thechocolate production was dened according tode Muijnck (2005).Since it is not relevant for the purpose of our study, no dataregarding retailer and consumers are included in the case study.The number of actors and amounts of product handled can be seeninTable 1.

Fig. 1. Supply chain of the case study.

Table 1

Overview of simulation model parameters, values or range of value and parametertype.

Simulation model parameter Value/range of values

Parametertype

Chocolate production process

Waste from cleaning 12.9% PProcessing batch size 1600e5000 kg PCocoa liquor extracted from nibs 87.1% PCocoa liquor to mixing process 62% PCocoa liquor to pressing process 38% PCocoa butter extracted 49% PCocoa liquor (fraction) 0.40 PCocoa butter (fraction) 0.12 PSugar (fraction) 0.47 PLecithin and others (fraction) 0.01 PPackaging size 0.1 kg PSupply chain

Cocoa beans produced per cocoa farmer 1000 kg C

Number of cocoa farmers 630 PNumber of local buying stations 12 PNumber of cocoa farmers per

local buying stations52e53 P

Cocoa beans per local buying station 52000e53000 kg PNumber of cocoa exporters 5 CNumber of local buying stations

per cocoa exporters2e3 C

Cocoa beans per cocoa exporter 104000e159000 kg PBatch size of the cocoa exporter 10000e18000 kg UNumber of batches per cocoa exporter 5e16 UTotal cocoa beans received/chocolate

manufacturer630000 kg P

Number of chocolate manufacturers 1 CTotal chocolate produced 740812.94 kg PNumber of packages 7408129.37 P

C: constant parameter; P: predetermined parameter; and U: uncertain parameter.

R. Saltini, R. Akkerman / Food Control 23 (2012) 221e226222


3/6

3.2. Simulation model

In order to analyze the impact of traceability improvements onthe consequences of a safety crisis we simulated recalls in the casestudy. To do this, we developed a spreadsheet simulation modelimplemented in Visual Basic for Microsoft Excel. These types ofsimulation models are often preferable because of the softwareusability and availability (Akkerman & van Donk, 2008). Thesimulation model is based on a series of parameters that are clas-sied as constant (C), predetermined (P) or uncertain (U). Constantparameters refer to parameters that remain unchanged, pre-determined to those that can actively be changed in the model, anduncertain to unknown parameters (see alsoAkkerman & van Donk,2010). For the unknown parameters, a probability distribution istypically dened and the simulation model randomly choosesa value for each simulation run. All data used for developing thesimulation model and their classication are shown inTable 1. Itshould be noted that wherever a range of values is denoted,a uniform distribution is used in the simulation.

In the chocolate production process the roasting processrepresents a key step for improving the microbiological conditionas well as for dening the aroma prole of the nal product (de

Muijnck, 2005). Safety and quality of the nished chocolatestrongly depend on the roasting process. Each batch of cocoa beansreceived by the chocolate manufacturer is usually split into severalprocessing batches, whose dimension depends on the capacity ofthe roasting equipment. Thus, each processing batch goes intoa specic roasting process. In this way, if a problem occurs to oneroasting process, only the nished chocolate produced with thatspecic roasting process will suffer the consequences of theproblem. The processing batch size is therefore a key (pre-determined) parameter in the simulation model, and an essentialplanning decision in practice.

3.3. Experimental design

3.3.1. Different production strategiesThe simulation model was designed to simulate the chocolate

production system for two different production strategies, onebased on production efciency (PS1) and one based on reducedbatch dispersion (PS2). In PS1 the maximum processing batch sizeis always used so that the equipment in the production stage isalways used at full capacity. Since the size of the cocoa bean batchesdelivered to the chocolate manufacturer is not necessarilya multiple of the processing batch size, some cocoa beans are mixedwith the next batch of cocoa beans. This results in having somebatches ofnished product produced from two different batches ofraw materials. Instead, PS2 focuses on reducing batch dispersion,where the chocolate manufacturer avoids mixing the differentbatches of cocoa beans. Here, some processing batches might be

smaller in size. As batch processes are involved, this results in somepartially unutilized processes in the chocolate production line, witha corresponding reduction in production efciency. On the otherhand, if a safety crisis occurs to a batch of raw materials, a PS2production strategy would lead to smaller recall sizes compared toPS1. A graphical illustration of both PS1 and PS2 can be seen inFig. 2.

In the remainder of this paper, production efciency ismeasured by the number of processing batches because:

The number of processing batches equals the number of timesa roasting process is performedand the duration of the roastingprocess depends on the roasting grade desired, not on theamount of nibs processed into the equipment (Jinap, Rosli,

Russly, & Nordin, 1998; de Muijnck, 2005). Therefore less

processing batches mean less time needed for roasting, witha constant number of equipments; or less equipments needed,

with a constant processing time required. Thus, less processingbatches lead to a higher efciency.

Smaller batch sizes (also meaning more batches when pro-cessing a constant raw material amount) were found by otherauthors to lead to an increase in production setup times andcosts, resulting in losses of production efciency (Dabbene &Gay, 2011; Dupuy et al., 2005; Rong & Grunow, 2010; Wang,Li, & OBrien, 2009).

3.3.2. Different traceability systems

The simulation model includes a basic traceability system (TS0),and two improved traceability systems (TS and TS). TS0 fullsthe European law regarding traceability, thus the actors involved in

the supply chain and located within the European borders followthe one step back-one step forward approach required by law(European Commission, 2002). That is, the nished chocolateproduct is traceable from the supermarket, to the chocolatemanufacturer, to the cocoa exporter. As the cocoa exporter isassumed to be located outside Europe it is not possible to trace thecocoa beans further in the supply chain.

TS is an extension of TS0, where the local buying stations,when buying the cocoa beans from the cocoa farmers, mark allcocoa bags with a unique code and the buying date. Also, when thecocoa exporter buys the cocoa and mixes bags from different localbuying stations, the original bags remain. This means the cocoa isdelivered to the chocolate manufacturer in the original bags withthe code of the local buying station. These codes are then registeredso that the nished chocolate can be traced up to the local buyingstation.

Fig. 2. Left: PS1 production strategy. Raw material batches are mixed in order to utilize

the maximum processing batch capacity. Right: PS2 production strategy, where raw

material batches are never mixed.

Fig. 3. Thegure shows up to where a chocolate bar is traceable when using TS0, TS

and TS

.

R. Saltini, R. Akkerman / Food Control 23 (2012) 221e226 223


4/6

TS extends TS with the addition that the cocoa farmers,when packing the cocoa beans, already mark all bags with uniquecodes and date. In this case the nished chocolate is traceable up tothe individual cocoa farmer. Alternatively, the local buying stationcould mark the bags at arrival, with the information of the farmerdelivering the beans.

InFig. 3the different traceability systems can be seen.

3.3.3. Different product recalls

The simulation model is able to simulate two possible foodcrises and corresponding recalls (R1 and R2) that could occur in thecase study supply chain. R1 simulates the product recall in case ofa contamination of the cocoa beans, which could be a chemicalcontamination while farming, fermenting or drying. In this case allchocolate bars produced with cocoa beans from a certain cocoafarmer need to be recalled. R2 simulates the product recall in caseof a contamination of a processing batch, which could be caused bya problem in a roasting process. In this case all chocolate barsproduced in a certain roasting process need to be recalled. Thesimulation models allow to run single and multiple simulations.Due to the importance of the roasting process it is also possible torun single or multiple simulations automatically for different pro-

cessing batch sizes. For this paper, we simulated the food scares fora range of processing batch sizes between 1,600 kg and 5,000 kg(every multiple of 200 kg). Each of the sizes is then run multipletimes, while information such as number of runs, processing batchsize, recall size and number of processing batches (which reectsthe production efciency) is registered, and average results can bedetermined.

4. Results and discussion

4.1. Comparing different production strategies

In order to compare the production strategy based on produc-

tion efciency (PS1) to the production strategy based on reducedbatch dispersion (PS2) the simulation model has been run fordifferent processing batch sizes. The average results are shown inFig. 4as the difference in percentages between the two production

strategies (values of PS1 represent the 100%) in ( i) productionefciency, (ii) recall size in case of contamination of the rawmaterials (R1) and (iii) recall size in case of contamination ofa processing batch size (R2).

PS2 is a production strategy where the batches of raw materialsare not mixed. Thus, the last processing batch from a raw materialbatch might not fully occupy the batch processing equipment. Thatis, the production equipment that processes this smaller batch willonly be partially utilized. The actual overall utilization thereforedepends on the processing batch size b and the raw materials batchsize n. Combined, these two factors lead to a required number ofprocessing batchesr(withr n/b) and a utilization of

u r

QrS100% (1)

whereQrSdenotes the smallest integer larger than or equal tor. Forsituations in which the processing batch size is xed (as can beexpected in industry), but raw material batch size varies, this leadsto different expected utilizations for each possible processing batchsizeb:

ub

1jNjXiN

riQriS (2)

whereriis the required number of batches for raw material batchsize ni, i N an index representing the different possible rawmaterial batch sizes used in the simulation (based on the uniformdistribution discussed in Section 2.2), and rNr the number ofelements in setN. The expected process utilization valueubfor theprocessing batch sizes simulated in our case study when using PS2can be seen inFig. 5.

Fig. 5shows that, by simulating a large number of possible rawmaterial batch sizes, and thus a large setN, the expected utilizationubvaries signicantly. When producing with PS1, the raw materialbatches are mixed, always processing with fully occupied batchprocessing equipment, thus reaching high utilization in the

production processes. Therefore, the number of processing batchesfor PS1, representing production efciency, has a decreasing trendfor an increasing processing batch size. As the decrease in thenumber of processing batches for PS2 is not constant, thepercentage difference between the two production strategies is notincreasing at a constant rate in Fig. 4. This is most visible at pro-cessing batch sizes around 4,400 kg, but smaller incontinuities canbe noticed around 2,800 kg and 3,600 kg. A chocolate manufacturerthat produces with a PS2 production strategy should therefore takethe expected utilization valueubas calculated in(2)and illustrated

Fig. 4. Difference between two production strategies in terms of production efciency

and recall sizes.

Fig. 5. Utilization of the roasting process (ub) at the different processing batch sizes (N)

simulated for PS2.



5/6

inFig. 5into account when deciding on the size of the processingbatches. Obviously, the recall size in case of contamination ofa processing batch (R2) strongly depends on the processing batchsize. Thus, this variable unutilized fraction of the process results ina variable difference in recall size between PS1 and PS2, which isreected in the peaks present in the R2 results curve in Fig. 4. Therecall size in case of contamination of the raw material (R1) is alsoaffected by this variable difference, but since this type of recallconsists of multiple batches ofnished product, this has a relativelysmall inuence on the results, and cannot be seen inFig. 4.

Overall, Fig. 4 shows that for all processing batch sizes therelative difference in production efciency ranges between 7 and22%, the relative difference in recall size for R1 between 0.5 and 4%,and for R2 between 6 and 16%. In general, when comparing PS1 toPS2, the difference of both production efciency and recall sizesbetween PS1 and PS2 increases when increasing processing batchsizes, due to the decreasing utilization and the increased size ofmixed batches.

When comparing the benets in increasedproduction efciencyof PS1 to the benets in reduced recall sizes of PS2, we did notinclude the probability that a product recall occurs. A product recallobviously does not occur for every batch produced, while benets

in production efciency inuence all produced batches. Therefore,when deciding whether adopting a new production strategy wouldbe benecial, the probability that a product recall occurs must beassessed and included in the decision-making process. However,even without this probability, the results obtained in this papershow that the effects in terms of production efciency in PS1 aresignicant. This suggests that adopting a production strategyfocused on reduced batch dispersion (such as our PS2 strategy)might often not be economically feasible. However, it should alsobe taken into account that the efciency losses and the amount oflost product are only part of the economical effects. It is well knownthat in case a product recall occurs, food companies also incur costsrelated to organizing the recall, bad publicity and damage to thereputation of the brand (Onyango et al. 2007). If these aspects were

quantied and taken into account, reducing recall sizes would havea bigger impact and PS2 might also become economically feasible.Quantifying the losses caused by bad publicity and damage to thereputation of a brand in case of product recall is outside the scope ofour paper.

4.2. Inuence of different traceability systems on product recall size

When constructing the case study it was seen that currently nofull traceability is in use in the supply chain of chocolate. This isprobably due to the fact that cocoa is farmed in non-Europeancountries, where there is no legislation obliging the actorsinvolved in a food supply chain to have traceability systems. TheEuropean legislation states that within Europe, food companies

must be able to identify immediate suppliers and customers ofa specic product (European Commission, 2002). This legislationguarantees full traceability within the European borders, buttraceability is lost when part of the supply chain goes outside theseborders. In the example case study constructed in this paper it wasseen that cocoa could be traced up to the local exporter, which,since located outside Europe, is not obligated to comply with theEuropean law on traceability and therefore does not record anyfurthertraceability information.

In our study the benets of having a traceability system that ismore accurate than what the law requires are quantiedin termsofreduction of product recalls. In order to propose realisticimprovements of the traceability system of the case study supplychain, TS and TS were designed taking into consideration that

most of the cocoa farmers would not have the resources to invest in

technologies, therefore these improvements imply very lowinvestments. In order to analyze the inuence of these differenttraceability systems (TS0, TSand TS) on the recall size of thenished product in case of contamination of the raw materials, weagain used the simulation tool. In order to have an accurate resultthe simulation model has been run 100 times for each processingbatch size and data are shown inFig. 6as the average of the aver-ages of the results of all roasting equipment sizes (kg).

The results show an average recall size of 1,608,719 units ofproduct for TS0, an average recall size of 714,584 units of productfor TSand an average recall size of 55,789 units of product forTS. Nostandarddeviation is shown in the Fig. 6 because it would

show a variation which is mainly caused by the different roastingequipment sizes, thus giving a false sense of variability. However,acceptable standard deviations were seen when analyzing theunderlying simulation results for each of the equipment sizes.These results show that in the case study supply chain, in case ofcontamination of the raw materials the adoption of TS wouldreduce the recall size of approximately 55% and the adoptiona TSwould reduce the recall size of approximately 96%. Again,the probability that a recall occurs would need to be considered to

justify the investment needed to set up such traceability systems.

5. Conclusions

This paper outlines and quanties the benets of traceabilityimprovements in the production system as well as in the entiresupply chain of chocolate. In the literature, the benets of trace-ability are largely discussed, but authors rarely quantify them. Webelieve that quantitative results on real or hypothetical improve-ments present a signicant help to food industries in the decision toimprove their traceability systems and how to do this.

We found that adopting a production strategy based on a lowbatch dispersion strategy would lead to a reduction of potentialproduct recall sizes between 0.5 and 4% in case of contamination ofthe raw materials and between 6 and 16% in case of contaminationof a processing batch size. On the other hand, it would also lead toa reduction of the overall production efciency between 7 and 22%.Wealso found that improving traceability in the entire supplychain

of chocolate could result in reduction of the magnitude of a recall in

Fig. 6. Recall size when simulating a food scare caused by contamination of the raw

material with different traceability systems.

R. Saltini, R. Akkerman / Food Control 23 (2012) 221e226 225


6/6

case of contamination of the raw material of 55% or 96%, dependingon the type of improvement.

It is crucial for the industry to take these results into accountwhen planning production improvements. These results show thatthe biggest benet might be achieved when improving traceabilitythrough the entire supply chain, more than when improving it ona single production system. Our results also show that adoptinga production strategy focused only on the reduction of batchdispersion would result in reduced losses in case of product recall,but it might not always be economically feasible when takingproduction efciency into consideration. In this paper we also usean expected process utilization value ub that can be used forassessing the optimal processing batch size. Especially for largerbatch sizes, this has a signicant effect on the realized productionefciency, but also on the size of potential recalls. Taking this valueinto consideration is therefore crucial for reducing batch dispersionand minimizing losses in production efciency.

In the future, a specic legislation for contaminants in cocoabeans will be enacted (Codex Alimentarius commission, 2007).Thus, increasing the control on the supply chain and improve cocoatraceability will even be a requirement. Therefore, identifying andquantifying the additional benets of a system that in the future

will anyway need to be improved might be of high interest for thechocolate as well as for other food industries. Here, we focus on thebenets of traceability implementations in terms of safety crisesand production efciency. According toGolan et al. (2004), trace-ability systems can handle an extended amount of information,which could represent a signicant added value of traceabilityimprovements. The characteristics of the nal chocolate productstrongly depend on the processes done at the very beginning of thesupply chain. Studies show that each cocoa farmer differs in theway he handles and processes the cocoa beans (Lain, 2001), Forinstance, avor precursors, such as free amino acids and reducingsugars, are formed during fermentation done by the farmers, andare transformed into avor compounds during the roasting of thecocoa beans. Theseavor compounds are subsequently responsible

for the avour prole of the nished chocolate (e.g.de Brito et al.,2001; Jinap et al., 1998; Ramli, Hassan, Said, Samsudin, & Idris,2006). Detailed information on product conditions and the way itwas farmed could be used to optimize processing parameters, suchas roasting temperature and cocoa bean selection. Further researchmust however be done to analyze these possibilities.

In conclusion, this paper shows that traceability systems arehighly valuable tools from which, if well designed and well inte-grated with the production operations, food producers can benetin terms of safety issues as well as in the daily operational activities.

References

Akkerman, R., Farahani, P., & Grunow, M. (2010). Quality, safety and sustainability infood distribution: a review of quantitative operations management approachesand challenges. OR Spectrum e Quantitative Approaches in Management, 32(4),863e904.

Akkerman, R., & van Donk, D. P. (2010). Balancing enviromental and economicperformance in the food processing industry. International Journal of Entrepre-neurship and Innovation Management, 11(3), 330e340.

Akkerman, R., & van Donk, D. P. (2008). Development and application of a decisionsupport tool for reduction of product losses in the food-processing industry.

Journal of Cleaner Production, 16, 335e342.Alfaro, J. A., & Rbade, L. A. (2009). Traceability as a strategic tool to improve

inventory management: a case study in the food industry. International Journalof Production Economics, 118(1), 104e110.

Canavari, M., Centonze, R., Hingley, M., & Spadoni, R. (2010). Traceability as part ofcompetitive strategy in the fruit supply chain. British Food Journal, 112(2),171e186.

ChocoSuisse, Association of Swiss Chocolate Manufactures. (2009). The SwissChocolate Industry in 2008. ChocoSuisse press release 09.02.2009.

Codex Alimentarius commission. (2007). Discussion Paper on Ochratoxin A inCocoa. Joint FAO/WHO Food Standards Programme Codex Committee onContaminants in Foods, pp. 11e14.

Dabbene, F., & Gay, P. (2011). Food traceability systems: performance evaluation andoptimization.Computers and Electronics in Agriculture, 75(1), 139e146.

de Brito, E. S., Garcia, N. H., Gallao, M. I., Cortelazzo, A. L., Fevereiro, P. S., &Braga, M. R. (2001). Structural and chemical changes in cocoa (Theobromacacao L) during fermentation, drying and roasting. Journal of the Science of Foodand Agriculture, 81, 281e288.

de Muijnck, L. (2005). Cocoa. In C. Onwulata (Ed.),Encapsulated and powdered foods(pp. 452e473). Boca Raton: Taylor & Francis Group, LLC.

Dupuy, C., Botta-Genoulaz, V., & Guinet, A. (2005). Batch dispersion model tooptimise traceability in food industry. Journal of Food Engineering, 70(3),333e339.

European Parliament and Council. (2002). General principles and requirements ofFood Law. Regulation (EC) No 178/2002. Ofcial Journal of the EuropeanCommunity, L31/1eL31/24.

Golan, E., Krissoff, B., Kuchler, F., Calvin, L., Nelson, K., & Price, G. (2004). Traceabilityin the US food supply: economic theory and industry studies. Agricultural economicreport number 830. Washington, DC: Economic Research Service, US Depart-ment of Agriculture.

Jinap, S., Rosli, W. I. W., Russly, A. R., & Nordin, L. M. (1998). Effect of roasting timeand temperature on volatile component proles during nib roasting of cocoabeans (Theobroma cacao). Journal of the Science of Food and Agriculture, 77(4),441e448.

Karlsen, K. M., Donnelly, K. A .-M., & Olsen, P. (2011). Granularity and its importancefor traceability in farmed salmon supply chain. Journal of Food Engineering,102(1), 1e8.

Kimball, A. M., & Taneda, K. (2004). A new method for assessing the impact ofemerging infections on global trade. Revue scientique et technique e Ofceinternational des pizooties, 23(3), 753e760.

Lain, K. (2001). Survey of farming practices on cocoa farms in Cte dIvoire. FieldStudy Report, PACCC/ICCO/Industry Project for the Improvement of CocoaQuality in Cte dIvoire.

Mai, N., Bogason, S. D., Arason, S., rnason, S. V., & Matthasson, T. G. (2010). Benetsof traceability in sh supply chains e case studies. British Food Journal, 112(9),976e1002.

Onyango, B., Miljkovic, D., Hallman, W., Nganje, W., Condry, S., & Cuite, C. (2008).Food recalls and food safety perceptions: the September 2006 spinach recallcase. Journal of Agribusiness, 26(1), 77e98.

Ramli, N., Hassan, O., Said, M., Samsudin, W., & Idris, N. A. (2006). Inuence ofroasting conditions on volatile avor of roasted malaysian cocoa beans. Journal

of Food Processing and Preservation, 30(3), 280e298.Regattieri, A., Gamberi, M., & Manzini, R. (2007). Traceability of food products:general framework and experimental evidence. Journal of Food Engineering,81(2), 347e356.

Rong, A., & Grunow, M. (2010). A methodology for controlling dispersion in foodproduction and distribution. OR Spectrum e Quantitative Approaches inManagement, 32(4), 957e978.

Ruiz-Garcia, L., Steinberger, G., & Rothmund, M. (2010). A model and prototypeimplementation for tracking and tracing agricultural batch products along thefood chain. Food Control, 21(2), 112e121.

Sarig, Y. (2003). Traceability of food products. CIGR Journal of Scientic Research andDevelopments, 5(12), 54e65.

Thakur, M., Wang, L., & Hurburgh, C. R. (2010). A multi-objective optimizationapproach to balancing cost and traceability in bulk grain handling. Journal ofFood Engineering, 101(2), 193e200.

Trautman, D., Goddard, E. W., & Nilsson, T. (2008). Traceability e a literature review,Vol. 08-02. Edmonton, Canada: Department of Rural Economy. Faculty of Agri-culture & Forestry, and Home Economics. University of Alberta, 1e151.

Viaene, J., & Verbeke, W. (1998). Traceability as a key instrument towards supply

chain and quality management in the Belgian poultry meat chain.Supply ChainManagement:An International Journal, 3(3), 139e141.

Wang, X., Li, D., & Li, L. (2009). Adding value of food traceability to the business:a supply chain management approach. International Journal of Services Opera-tions and Informatics, 4(3), 232e257.

Wang, X., Li, D., & OBrien, C. (2009). Optimisation of traceability and operationsplanning: an integrated model for perishable food production. International

Journal of Production Research, 47(11), 2865e2886.Wilcock, A., Ball, B., & Fajumo, A. (2011). Effective implementation of food safety

initiatives: Managers, food safety coordinators and production workersperspectives.Food Control, 22(1), 27e33.

World Health Organization. (2002). Bovine spongiform encephalopathy. Fact sheet,Vol. 113.


01 trazabilidad en alimentos

Documents