La web de esic.edu utiliza cookies para obtener datos estadísticos de la navegación de sus usuarios. Si continuas navegando, consideramos que aceptas su uso. Puedes ampliar esta información aquí. ►►Aceptar
JEL Code: O13, QO1, Q12.
Isanta-Muñoz, F., Moyano-Salvago, R., Villarroel-Molina, O., García de Tena, A., and Barba-Capote, C. (2020). Process management in the traceability system LeTrA Q of goat and sheep milk in Andalusia, Spain, Esic Market Economics and Business Journal, 51(2), 341-359. Doi: 10.7200/esicm.166.0512.3Download in PDF Format
The various food crises that the European Union have experienced in the late 20th century, as, for example, the bovine spongiform encephalopathy or “mad cow disease” (1996) or the use of dioxins in broiler chicken feed manufacturing (1999), confirm the lack of a system to collect and analyse data at the Community level on the food supply chain. This fact has forced the establishment of a data collection and analysis system by the health authorities. The implementation of a comprehensive traceability process in food and animal feeding companies allowed specific and precise product recalls to be carried out and informed consumers and control officials, thus avoiding further disruption in case of food security issues (Isanta et al., 2018; UE, 2002). Accordingly, the Member States undertook procedures for food crisis management and prevention. These organizational procedures should allow better coordination of efforts and determine the appropriate measures, depending on the best scientific information. Therefore, the revised processes should consider the responsibilities of the Competent Authorities and provide their technical and scientific support as a recommendation in the case of a food crisis. The European Union (EU) established in 2002 Regulation 178/2002, strict regulations on food hygiene, locating the responsibility of safe food production on the different agents that intervene throughout the food chain (EU, 2002). For several years, specific regulations on the hygiene of food coming animal origin have been developed, which allow the traceability of food at all stages of production, transformation, and distribution (MAPA, 2004; UE, 2004a, 2004b; MAPA, 2008). Therefore, since 2004, the identification and registration of agents were regulated, as well as the establishments, containers, and movements of raw milk (MPR, 2011), and it established a computer traceability management system based on barcodes named LeTrA Q in 2012 (MAPA, 2011). Production decision-making requires a systemic and dynamic approach, the interactions between the different elements of the system are contemplated, and the company is organized in a market context (Vilaboa-Arroniz, 2013; Villarroel- Molina et al., 2019). Management implies knowledge of the industry itself, and the human factor that integrates it (De-Pablo-Heredero and Blanco-Jiménez, 2013). Besides, it is materialized through decision-making and activates the functioning of planning processes, which generate data that is subsequently analysed (de Pablos-Heredero et al., 2012). Process management is the set of actions aimed at optimizing work in the company and pursues the improvement of an organization’s activities by identifying, describing, registering, selecting and, continuously improving processes (De-Pablos-Heredero and Blanco-Jiménez, 2013; Isanta et al., 2018). Therefore, the study of process management has proven to be useful for improving operating procedures that further reduce the risk of a food crisis, and favour the implementation of preventive and corrective measures, which contribute to guaranteeing food security. Consequently, this paper aims to analyse the process management in the traceability system LeTrA Q of goat and sheep milk, and thus contribute to the knowledge of the business value chain.
The dairy sector in Spain and Andalusia
In the European Union, with 4% of the total, Spain is the 7th country producing cow milk. The country’s production ascends to the first positions over sheep milk, Spain occupies the first position with 17% of the total, and goat milk, with which the country reached 2nd place and 22% of the total (MAPA, 2019). The dairy sector is strategic in the Spanish agri-food industry due to its economic importance and contribution to maintaining the living condition of the rural population. An 88.9% of the total milk produced by Spanish farmers comes cows, 5.7% sheep, and 5.4% goats (FEGA, 2018). Within the livestock sector, the dairy sector is the second in importance, generating nearly 12,000 million euros a year throughout the entire dairy production and transformation chain and employing about 80,000 people, thus favouring other sectors of the economy, as the logistics industry (MAPA, 2019). According to the goat census in 2018, Spain occupies the second position in the European Union after Greece. Specifically, although Greece leads the production (32% of the European goat population), Spain has around 3 million heads, which corresponds to 25% of the total. On the other hand, Romania and France occupy the third and fourth positions, respectively (MAPA, 2019). In 2016, the quantity of goat milk produced in Spain farms was approximately 507 million litters. In the country, Andalusia is the main producer (40.7%), followed by Castilla La Mancha (17.6%), Canarias (12.5%), Extremadura (7.8%), and Castilla y León (5.4%). The remaining Autonomous Communities represent less than 2% of the total production (MAPA, 2019). Furthermore, Castilla y León (42%), and Castilla La Mancha (37%) are the two major producers of dairy sheep in Spain, according to the breeding females’ census of 2018. Besides, Extremadura ranks third (6.5%), followed by Navarra (3.9%), and País Vasco (3.3%). However, the Andalusian sheep industry (minority within the sector at the state level), amounts to 1.8% of the national total, which is slightly more than 40,000 breeding females of the approximately 2,255,000 sheep milk registered in Spain (MAPA, 2019). The milk sheep Spanish production in 2016 was 544 million litters, prominent in Castilla y León (53.7%), and Castilla La Mancha (33%), followed far behind by Navarra (2.84%), and Madrid (2.73%). Nevertheless, Andalusia, Extremadura, and País Vasco showed low values, under 2%. The production in the remaining Autonomous Communities contributes less to the dairy sheep industry (MAPA, 2019).
The structure of the dairy goat sector in Andalusia
Concerning the goat milk-producing females, the majority are located in Malaga (22%), Seville (17.34%), Almería (15.54%) and Granada (15.41%), followed by Cádiz (12%), and with less population, Córdoba (6.24%), Jaen (5.79%) and Huelva (5.77%) as it is shown in Figure 1. Figure 1. Distribution map of dairy goat farms in Andalusia, according to density N.º of farms 0 1-25 26-50 51-75 76-100 > 100 Source: Adapted SIGGAN 2016.
The structure of the dairy sheep sector in Andalusia
In Andalusia, the main sheep milk producers are Huelva (28%) and Córdoba (21%). Huelva has a business-oriented production system, while the production in Cordoba corresponds mainly to a phenomenon of productive transformation due to the dairy cattle sector restructuring in El Valle de Los Pedroches. Apart this, Seville, Cádiz, and Málaga have traditional livestock production systems in mountain areas and are the second-largest source of female milking sheep with 12%, 11%, and 10%, respectively. Finally, Almería (7.7%) and Granada (5.9%) are the provinces with the least representation in the sector. Moreover, the total number of sheep farms with registered activity in milk production amounts to 209 units (CAGPDS, 2019), which are distributed geographically in the provinces of Malaga (46), Cádiz (40), Cordoba (30), Granada (30), Sevilla (33), Almeria (22), Granada (30), Huelva (2) and Jaen (6).
Quality control system: LeTrA Q
LeTrA Q is an information system in a web environment that allows the registration and identification of agents, establishments, and containers that are part of the dairy sector. This system provides information on raw milk movements, as well as the results obtained the analysis of the samples of raw milk intended for human consumption, taken both on the farm and when unloaded at the dairy centre (MAPA, 2018). The main objective is to improve the control of the traceability and quality of raw milk, the transparency of the dairy sector the producer farm to the processing industry, and guarantee food security. For this, the Traceability Module of the “LeTrA Q database” was created, a software application the agents and containers in the dairy sector are registered. In this computer application, the managers of the dairy centres record the movements among containers, the moment the raw milk leaves the producing farm until it reaches the processing centre. It also guarantees compliance with the relevant quality requirements established in the community regulations on food hygiene. Subsequently, a module was developed that allows the high-quality management of raw cow’s milk samples corresponding to mandatory controls and livestock movements. This enables the actor’s registration that participate throughout the milk supply chain. LeTrA Q is part of the research promoted by the Ministry of Agriculture, Fisheries, and Food (MAPA), to improve the quality control and raw milk traceability; to enhance he dairy sector transparency the farm to the processing centre. Besides, this initiative meets the consumer’s demand for a higher guarantee in terms of food security (Isanta et al., 2018). This database was created through Royal Decree 319/2015 of April 24 and is channelled by the FEGA-Spanish foundation of Agriculture Guarantee. - depending on the Ministry of Agriculture, Fisheries, and Food (FEGA, 2018). It has a unified information system for the dairy sector (INFOLAC) and includes information on mandatory declarations of raw milk deliveries and the contracts among buyers and producers in Spain.
To conduct the analysis 84,484 goat milk samples have been used. These samples belong to the self-inspection system of the 628 farms distributed in Almeria (282), Malaga (211), and Córdoba (135). Apart this, we have collected 7,507 sheep milk samples 53 farms across different provinces as Cordoba (28), Cadiz (18), Malaga (4), Jaen (2), and Huelva (1). Furthermore, we used data 5 laboratories, (A) Laboratorio Interprofesional Lechero de Cantabria -LILC-, (B) Laboratorio Interprofesional Lechero de Castilla La Mancha –LILCAM-, (C) Laboratorio Interprofesional Lechero de Castilla y León -LILCYL, (D) Centro Tecnológico Agroalimentario de Pozoblanco -CICAP-, and (E) Laboratori d’Anàlisi de LLet - de la Universidad Politécnica de Valencia -LICOCAL-, which make up the database of the raw goat milk traceability system named LeTrA Q (MAPA, 2011). This information was analysed its creation in 2012 to 2016. Here, Spanish acronyms have been kept unchanged. The process management qualitative variables considered are shown in Table 1. Table 1. List of variables analysed Variable Acronym Category Acronym Subcategory Acronym Sample status EM Sample received and analysed AN Sample received and not analysed NA Sample identification problem IM Sample removal/transporting problem RM Incidence of sample collection TM Control problems between sample collection and samples receiving dates CM Visible defects in the sample DM Defective container removal ED Repeated sample MR Sample not applicable to analyse NC Variable Acronym Category Acronym Subcategory Acronym Sample result RM Samples in reserve ER Valid sample VA Incomplete valid sample VI Analysis process interruption IP Very high bacteriology suggesting poorly taken or poorly preserved samples BA Control problems between sample collection and samples receiving dates CM Fat and/or protein result out of range RF Sample not applicable to analyse NC Rejected samples RE Error in the analysis of process FA Visible sample alteration AV Very high bacteriology suggesting poorly taken or poorly preserved samples BA Duplicate samples MD Fat and/or protein result out of range RF * Spanish acronyms have been kept unchanged. For the process management, the following quantitative variables have been considered, time the sample collection to the reception in the laboratory (TT-R), time sample reception to its analysis (TR-A), and the total time the sample collection to its analysis (TT-A). The time is expressed in hours. Firstly, a descriptive statistical analysis of the qualitative variables studied was performed: (ME), considering the two possible variants (AN and NA). Besides, (RM), with four options (VA, VI, EN, and RE) was considered. Subsequently, non-parametric analysis of variance was carried out (Kruskal-Wallis test) for the qualitative variables EM and RM, using the province as a variation factor. Then, both the homogeneity and the Kolmogorov-Smirnow tests, have been performed. Secondly, for the quantitative process management variables (TT-R, TR-A, and TT-A), a descriptive statistical analysis has been performed. Measures of statistical dispersion as the quartiles, standard deviation, standard error of the mean and coefficient of variation, considering the mean as the principal measure of central tendency, have been used. Subsequently, for the quantitative variables, a univariate analysis of variance has been performed by taking into account the year and the laboratory as a fixed-effect. Finally, for the comparison of means, the Duncan test has been applied. The software version 10 has been used (StatSoft, 2011).
Results and Discussion
Table 2 shows the descriptive statistics of the qualitative process management variables used in the study. Findings indicated that of the 84,484 goat milk samples, 84,478 were AN and only 6 NA. We have also found that the NA cases corresponded to the NC type, while causality among IM, RM, TM, CM, DM, ED or MR, was not evidenced. Furthermore, of the 84,478 samples analysed (AN), 45.380 were valid and analysed (VA) (53.83 ± 0.002%), while 37,776 were incomplete valid sample (VI) (44.81 ± 0.002%), 538 reserve samples (ER), and 602 correspond to rejected samples (RE) (0.64 ± 0.003% and 0.71 ± 0.003%, respectively). The results have evidenced that the VI cases corresponded to the NC typology, while Monitoring bulk milk quality by an integral traceability system of milk IP, BA, CM and RF. The results have also shown that RE cases correspond to the FR typology, though cases falling into the FA, AV, BA or MD categories were not found. Table 2. Descriptive statistics of the qualitative process management variables in goat milk EM N PM ± EE (%) RM N PM ± EE (%) AN 84,478 99.99±2.9 10-5 VA 45,380 53.83±0.002 VI 37,776 44.81±0.002 ER 538 0.64±0.0003 RE 602 0.71±0.0003 NA 6 0.01±2.9 10-5 EM: sample status; AN: analysed sample; NA: Sample received and not analysed; RM: Sample results; VA: valid sample; VI: incomplete valid sample; ER: sample in reserve, RE: rejected sample; N: number of data; PM: average ratio; EE: standard error of the average ratio. Table 3 shows the results for the comparative analysis of the qualitative process management variables in goat milk. In this case, the province has been used as a variation factor. Regarding EM, we have found statistical homogeneity among the different provinces, and highly significant differences for RM. For the VA value, Cordoba has shown the highest value, while Malaga showed the lowest. In both Cordoba and Malaga, the most frequent variant within RM was VA, though in Almería, it was VI. Table 3. Non-parametric analysis of variance and mean homogeneity test for goat milk variables Variables Almería Córdoba Málaga N % N % N % EM AN 46,689 99.99ª 21,547 99.99ª 16.242 99.99ª NA 3 0.01ª 2 0.01ª 1 0.01ª RM VA 18,673 39.99c 16,856 78.23ª 9.923 61.09b VI 27,794 59.53ª 4,178 19.39c 5.912 36.40b ER 81 0.17b 247 1.15ª 211 1.30ª RE 141 0.30b 266 1.23ª 196 1.21ª EM: sample status; AN: analysed sample; NA: Sample received and not analysed; RM: Sample results; VA: valid sample; VI: incomplete valid sample; ER: sample in reserve, RE: rejected sample; N: number of data; The same letter in the same row indicates homogeneity, while different letter in the same row indicates statistical differences (p<0.01). Concerning the handling and processing of samples in laboratories, the low values observed for AN and RE allow us to infer a correct operation of the Letra Q traceability system in the goat species. However, a high proportion of VI was found, which is attributed to the frequency of milk sampling. In this case, the sample collection frequency exceeded the fortnightly analysis established in the regulations (MAPA, 2004). This suggests a protocol oversizing. This situation involves additional costs, mainly related to the sample collection on the farm, transport logistics, as well as laboratory expenses (laboratory technicians and specific operating expenses). Therefore, an improvement is needed to optimize the cost of the traceability system LeTrA Q. The results obtained for the response times in the sample management are shown in table 4. Here, sample collection and sample analysis (TT-A) have shown the highest mean value (59.04 ± 31.92 in hours). However, the TT-R value is slightly above the limit established (48 hours) in the official regulations (MAPA, 2004). Table 4. Descriptive statistics of the qualitative process management variables Variables N Media Q1 Q3 SD Cv TT-R 37,764 50.16 24 72 35.52 0.71 TR-A 37,764 8.88 0.00 7.2 24.48 2.72 TT-A 37,764 59.04 24 72 31.92 0.54 TT-R: sample collection and sample reception timing in the laboratory; TR-A: sample reception and sample timing analysis in the laboratory; TT-A: sample collection and sample timing analysis in the laboratory; N: number of data; Q1: first quartile; Q3: third quartile; SD: standard deviation; Cv: variation coefficient. The comparative analysis of the process management variables reflected the existence of statistical differences for the laboratory factor (Table 5). Accordingly, the highest value for TT-R was 105.12 hours, found at the Laboratori d’Anàlisi de LLet – at the Universidad Politécnica de Valencia (E). Table 5. Analysis of variance for the process management variables, considering the laboratory as a variation factor Process management times Laboratories A B C D E TT-R 53.76c 74.64b 51.36c 2.4d 105.12ª TR-A 0.48c 0.24c 0.24c 58.56ª 16.08b TT-A 54.24d 74.88b 51.6d 60.96c 121.2ª (A) Laboratorio Interprofesional Lechero de Cantabria -LILC-; (B) Laboratorio Interprofesional Lechero de Castilla La Mancha –LILCAM-; (C) Laboratorio Interprofesional Lechero de Castilla y León -LILCYL; (D) Centro Tecnológico Agroalimentario de Pozoblanco -CICAP-; (E) Laboratori d’Anàlisi de LLet - de la Universidad Politécnica de Valencia -LICOCAL-; TT-R: sample collection and sample reception timing in the laboratory; TR-A: sample reception and sample timing analysis in the laboratory; TT-A: sample collection and sample timing analysis in the laboratory; The same letter in the same row indicates homogeneity, while different letter in the same row indicates statistical differences (p<0.01). Three highly homogeneous groups were found, regarding TR-A. The first of those corresponds to CICAP (D) and registered the highest time (58.56 hours) for this variable, the second corresponds to the milk testing laboratory of the Universidad Politécnica de Valencia (E), and the third group is made up of the other three laboratories. This latter group registered the fastest time of sample analysis. Finally, the variable TT-A showed very similar values to the TT-R, suggesting that the use of one of them is enough to improve this area of process management. Regarding the year as a variation factor, the univariate comparative analysis for the process management variables showed the existence of significant differences for all the variables (Table 6). Since the launch of the program in 2012 to 2015, an increase in TT-R values was observed, with a decrease in 2016. This may be due to the continuous increase of analysed samples and the adhesion of new farms to the LeTrA Q system. Different groupings were established under years analysed, except in 2013 and 2014, a single homogeneous group was formed. On the other hand, three highly homogeneous groups were found, regarding TR-A. The first of those corresponds to the year 2012 and registered the highest value of the period under study, the second corresponds to the year 2016, while the third group correspond to the years 2013, 2014, and 2015. Finally, the behaviour of TT-R variables has been very similar to those of the TR-A. Although they have been grouped into three different groups, as follows: 2012 (group 1), 2015 and 2016 (group 2), and 2013-2014 (group 3). Table 6. Analysis of variance for the process management variables, considering the year as a variation factor Process management times Year 2012 2013 2014 2015 2016 TT-R 30.48d 49.2b 49.92b 56.64ª 46.32c TR-A 34.8a 5.04c 6c 6c 16.32b TT-A 65.28a 54.24c 55.92c 62.64b 62.64b TT-R: Sample collection and sample reception timing in the laboratory; TR-A: Sample reception and sample timing analysis in the laboratory; TT-A: Sample collection and sample timing analysis in the laboratory; The same letter in the same row indicates homogeneity, while different letter in the same row indicates statistical differences (p<0.01). The results show the need to improve the logistics of collecting and transporting the samples to the laboratories, to reduce time. Besides, it would be necessary to have an interprofessional dairy laboratory in Andalusia developed by the Public Administrations, since Andalusia presents a high number of livestock farms. As an alternative, the existing laboratory CICAP (the only dairy laboratory located in Andalusia), could be expanded and improved. Since it has a strategic location in El Valle de Los Pedroches, an important region in the dairy industry.
Table 7 shows the descriptive statistics of the qualitative process management variables used in the study. Findings indicated that of the 7,507 sheep milk samples, just 1 was NA since it corresponds to NC samples. Apart this, causality among IM, RM, TM, CM, DM, ED or MR, was not evidenced, which confirms that the traceability system implemented is effective. Furthermore, of the 7,506 samples analysed (AN), 5,830 were valid analysed (VA), corresponding to (77.67 ± 0.54%) of the total, while 1,550 were incomplete valid sample (VI) (20.65 ± 0.53%), 79 were samples in reserve (ER), and 47 correspond to rejected samples (RE), 1.05 ± 0.01% and 0.63 ± 0.01%, respectively. The results have shown that the VI is divided into NC (1,547) and IP (8), while observations were not found regarding BA, CM and FR. On the other hand, the results have also shown that RE cases correspond to the FR typology, though cases falling into the FA, AV, BA or MD categories were not found. Finally, ER corresponds to those samples that are stored by the laboratories in custody as control samples. Table 7. Descriptive statistics of the qualitative process management variables for sheep milk EM N PM ± EE (%) RM N PM ± EE (%) AN 7,506 99.99±1.3 10-4 VA 5,830 77.67±0.54 VI 1,550 20.65±0.53 ER 79 1.05±0.01 RE 47 0.63±0.01 NA 1 0.01±1.3 10-4 EM: Sample status; AN: Analysed sample; NA: Sample received and not analysed; RM: Sample results; VA: Valid sample; VI: Incomplete valid sample; ER: Reserve sample, RE: Rejected sample; N: Number of data; PM: Average ratio; EE: Standard error of the average ratio. The results have shown low laboratory incidents (IP, RE), and an oversizing in the number of samples collected (NC), exceeding the minimum established by the regulation. This generates additional costs in the program that could be avoided to improve the farmers’ economy and reduce expenses in the public administrations. Table 8 shows the results for the comparative analysis by provinces of the process management qualitative variables in sheep milk. The RM variable was highly significant (p <0.001), obtaining the highest percentage of valid sample (VA), in each province. Milk samples with ER were found, in Cordoba, Jaen and Huelva; the percentages for this variable were 24.5%, 13.96% and 8.85%, respectively. Regarding NA, 9 samples were found, distributed among the provinces of Cordoba (6), Huelva (2), and Malaga (1). However, the sample status (EM) was not significant (p> 0.05). Therefore, it can be deduced that sample collection and management is carried out differently in each province, probably due to the implementation of different action protocols. The results for the response times in the sample management process are shown in table 9. These findings are significantly lower than the maximum limits established by the current regulation (MAPA, 2004). Table 8. Non-parametric analysis of variance and mean homogeneity test for sheep milk variables Variables Cadiz Cordoba Huelva Jaen Malaga N % N % N % N % N % EM AN 169 100.00a 5,656 100.00a 1,152 100.00a 394 100.0a 134 99.26a NA 0 0.00 6 0.00 2 0.00 0 0.00 1 0.74 RM VA 167 98.82ª 4,180 73.90c 1,028 89.24a,b 325 82.49b 130 96.30c VI 2 1.18 53 0.94 20 1.74 6 1.52 5 3.70 ER 0 0.00 1,386 24.50 102 8.85 55 13.96 0 0.00 RE 0 0.00 37 0.65 2 0.17 8 2.03 0 0.00 EM: Sample status; AN: Analysed sample; NA: Sample received and not analysed; RM: Sample results; VA: Valid sample; VI: Incomplete valid sample; ER: Sample in reserve, RE: Rejected sample; N: Number of data; The same letter in the same row indicates homogeneity, while different letter in the same row indicates statistical differences (p<0.01). Table 9. Descriptive statistics of the quantitative process management variables in sheep milk Variables N Media Min. Max. SD C.V. TT-R 5,830 30.48 0.00 720,00 43.68 143.07 TR-A 5,830 31.68 0.00 1008 40.08 126.58 TT-A 5,830 62.16 0.00 1080 42.48 68.36 TT-R: Sample collection and sample reception timing in the laboratory; TR-A: Sample reception and sample timing analysis in the laboratory; TT-A: Sample collection and sample timing analysis in the laboratory; N: Number of data; SD: Standard deviation; Min: Minimum value; Max: Maximum value; Cv: Variation coefficient. The comparative analysis of process management quantitative variables for the two factors considered was significant (p <0.005) and is shown in table 10. Furthermore, when considering the year as a variation factor, a progressive increase in the TT-R was found 2013 to 2016; with significant differences for each year, except in 2015 and 2016, a highly homogeneous group was found. On the other hand, the results for TR-A were very different those for the variable TT-R. In this case, the samples time of analysis was considerably reduced year after year, which indicates an improvement in logistics and process management. However, the TT-A was less informative since this variable is the sum of the previous times that showed antagonistic behaviour among them. Therefore, the proper functioning of the sheep milk control and traceability system in Andalusia is deduced. It should be noted that by implementing monitoring systems for milk samples during transport, such a system can be improved (De la Vara et al., 2018). Furthermore, the progressive increase in the TT-R 2012 to 2016 is due to the growth in the number of samples collected annually, without an increase in the number of technical staff. Finally, throughout the period analysed, a gradual decrease in the value of TR-A was evident, which indicates an adequate adaptation and improvement in laboratory management processes logistics, including the automation of the analyses. Regarding the laboratory, significant differences (p <0.05) were found for all the variables. A greater similarity was observed between the laboratories for TR-A and TT-A (with two and three homogeneity groups, respectively). Besides, TT-R was the variable with the greatest differences between the four laboratories, obtaining the lowest value (6) in the laboratory of Pozoblanco, in Cordoba. Although the samples take less time to reach Laboratory D, it resulted in the highest TR-A times. In contrast, both laboratories B and C resulted in a value of 0 for TR-A. These results are consistent with the farm’s location since the farms send the sample to the closest laboratory, which therefore reduces the TT-R. Apart this, the TR-A values increase when many samples are received; compared to remote laboratories that, because of their location, receive fewer samples and can analyse them almost immediately. The TT-R is different for each laboratory given the location of these in different Spanish regions (northern zone: 2, central zone 1, southern zone: 1). Therefore, the geographical distance between the laboratories and the farms varies the transport time of the samples. Besides, the TR-A is homogeneous among the three external analysis laboratories in Andalusia, which being interprofessional dairy laboratories, have a shorter execution time than the laboratories located in Andalusia. Response times are considered satisfactory and contribute to rapid decision-making, especially the TT-R, that is within the maximum period of 48 hours (h) established in the regulations (MAPA, 2004). Table 10. Analysis of variance for the process management variables in sheep milk, considering the laboratory and the year as a variation factor Laboratory Year A B C D 2012 2013 2014 2015 2016 TT-R 62.4b 69.36ª 43.92c 6d 8.4c 3.36d 26.16b 34.08ª 36ª TR-A 0.24b 0.00b 0.00b 58.56ª 52.32ª 53.28ª 35.28b 27.12c 27.6c TT-A 62.64b 69.36ª 43.92c 64.56b 60.72ab 56.64b 61.44ab 61.68ab 63.6ª TT-R: Sample collection and sample reception timing in the laboratory; TR-A: Sample reception and sample timing analysis in the laboratory; TT-A: Sample collection and sample timing analysis in the laboratory; A: Laboratorio Interprofesional Lechero de Cantabria -LILC-; B: Laboratorio Interprofesional Lechero de Castilla La Mancha –LILCAM-; C: Laboratorio Interprofesional Lechero de Castilla y León -LILCYL; D: Centro Tecnológico Agroalimentario de Pozoblanco -CICAP- y, E: Laboratori d’Anàlisi de LLet -Laboratorio de Análisis de Leche del Instituto de Ciencia y Tecnología Animal- de la Universidad Politécnica de Valencia -LICOCAL-; The same letter in the same row indicates homogeneity, while different letter in the same row indicates statistical differences (p<0.01). Therefore, differences in the dairy production system of goat and sheep have been found; both, for the physiological characteristics and the farming systems, among other factors. However, considering the LeTrA Q system, it is necessary to carry out a small comparative examination between the two species. On the other hand, different results were obtained in both species, for the variables of qualitative and quantitative process management. Therefore, the results in the sheep sector were better, since this sector presents a higher proportion of valid samples analysed with less time in the analysis of the samples; this is because the goat sector presents a larger dimension. Moreover, the traceability system has counts on with the infrastructure and logistics necessary for the implementation, execution and integration of the action protocol, required for its correct functioning (Brown, 2009). Therefore, this facilitates the positioning and commercialization of traceable milk and dairy products through a governmental control program, developed by industry associations or certified by other operators (Bai et al., 2013).
The analysed variables demonstrated their usefulness in the management of the traceability system LeTrA Q for raw goat and sheep milk in Andalusia. These variables generated dynamic and fluid information, which contributes to guaranteeing consumers food security by Public Administrations. Besides, this system constitutes a tool of great interest for the management and improvement in the information standards and business knowledge of the production and processing sector at a general level, and the quality parameters at the individual level. The proper functioning of the LeTrA Q system is the base to the consolidation of a self-control model for the different operators involved in the dairy sector value chain, which along with the quality and reliability of the information improve the strategic positioning of the dairy sector of the European Union within and outside its borders. The results for the process management variables suggest the existence of dysfunctions in logistics operations concerning the sending samples to the laboratories of analysis, given the average high value of TT-R. However, the results of the remaining variable confirm the proper functioning of the LeTrA Q traceability system, within the range of normal. The proper functioning of the LeTrA Q traceability system in Andalusia improve the consumer’s food security and constitutes a valuable tool to enhance the technical- economic management of farms.
Bai, J., Zhang, C., & Jiang, J., 2013, The role of certificate issuer on consumers’ willingness- to-pay for milk traceability in China. Agricultural Economics, 44(4-5), 537-544. doi: 10.1111/agec.12037 Brown, B. R., 2009, Maize to milk: An analysis of the traceability systems of bulk commodities (Undergraduate thesis). Iowa State University, Iowa, USA. https:// lib.dr.iastate.edu/etd/10768 CAGPDS [Consejería de Agricultura, Ganadería, Pesca y Desarrollo Sostenible], 2019, Caracterización del sector ovino y caprino en Andalucía. Recuperado de: http://ws128.juntadeandalucia.es/agriculturaypesca/portal/export/sites/default/comun/galerias/galeriaDescargas/cap/servicio-estadisticas/Estudios-e-informes/ ganaderia/ovino-caprino/ovino_caprino_2015.pdf De la Vara Martínez, J. A., García Higuera, A., Román Esteban, M., Romero Asensio, J., Carmona Delgado, M., Berruga, I., & Molina, A., 2018, Monitoring bulk milk quality by an integral traceability system of milk. Journal of Applied Animal Research, 46(1), 784-790. doi: 10.1080/09712119.2017.1403327 De Pablos-Heredero, C., López-Hermoso-Agius, J. J., Martín-Romo Romero, S., & Medina-Salgado, S., 2012, Organización y transformación de los sistemas de información en la empresa. ESIC Editorial. Madrid. España. De Pablos-Heredero, C., & Blanco-Jiménez, F., 2013, Los cien errores del emprendimiento. ESIC Editorial. Madrid. España. FEGA [Fondo Español de Garantía Agraria], 2018, Declaraciones obligatorias del sector ovino/caprino. Recuperado de: https://www.fega.es/ datos-campanas-clasificadas-por-actividad/actividad/ovino. Isanta, F., Barba, C., León, J. M., García de Tena, A., Angón, E., & Moyano, R., 2018, Análisis prospectivo del sistema de trazabilidad de la leche de oveja en el sur de España: resultados preliminares. Revista Científica FCV-LUZ. Vol. XXVIII (5); 360-368. https://produccioncientificaluz.org/index.php/cientifica/article/ view/29766/0 MAPA [Ministerio de Agricultura, Pesca y Alimentación], 2004, Real Decreto 217/2004, de 6 de febrero, por el que se regulan la identificación y registro de los agentes, establecimientos y contenedores que intervienen en el sector lácteo, y el registro de los movimientos de la leche. Núm. 43. Recuperado de: https://www. boe.es/buscar/pdf/2004/BOE-A-2004-3065-consolidado.pdf – 2008, Real Decreto 1728/2007, de 21 de diciembre, por el que se establece la normativa básica de control que deben cumplir los operadores del sector lácteo y se modifica el Real Decreto 217/2004, de 6 de febrero, por el que se regulan la identificación y registro de los agentes, establecimientos y contenedores que intervienen en el sector lácteo, y el registro de los movimientos de la leche. Núm. 15. Recuperado de: https://www.boe.es/buscar/pdf/2008/BOE-A-2008-823-consolidado. pdf 2011, Letra Q. Recuperado de: http://www.magrama.gob.es/es/ganaderia/temas/ trazabilidad-animal/letra-q/ – 2018, Letra Q. En línea: https://www.mapa.gob.es/es/ganaderia/temas/sanidad- animal-higiene-ganadera/higiene-de-la-produccion-primaria-ganadera/calidad- de-la-leche-letra-q/ (Fecha de consulta: 19/09/2018). – 2019, Información del sector ovino y caprino. Recuperado de: https:// www.mapa.gob.es/es/ganaderia/temas/produccion-y-mercados-ganaderos/ sectores-ganaderos/ovino-caprino/informacion-del-sector/ MPR [Ministerio de la Presidencia, Relaciones con las Cortes e Igualdad], 2011, Real Decreto 752/2011, de 27 de mayo, por el que se establece la normativa básica de control que deben cumplir los agentes del sector de leche cruda de oveja y cabra. Núm. 139. Recuperado de: https://www.boe.es/buscar/pdf/2011/ BOE-A-2011-9995-consolidado.pdf StatSoft, Inc., 2011, STATISTICA (data analysis software system), version 10. www. statsoft.com. Tulsa, USA. UE [Unión Europea], 2002, Reglamento (CE) n.º 178/2002 del Parlamento Europeo y del Consejo, de 28 de enero de 2002, por el que se establecen los principios y los requisitos generales de la legislación alimentaria, se crea la Autoridad Europea de Seguridad Alimentaria y se fijan procedimientos relativos a la seguridad alimentaria. Diario Oficial de la Unión Europea, 31 de 01.02.2002: 1-24. Recuperado de: https://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=CONSLEG:2002R0178:20080325:ES:PDF UE [Unión Europea], 2004a, Reglamento (CE) n.º 852/2004 del Parlamento Europeo y del Consejo de 29 de abril de 2004 relativo a la higiene de los productos alimenticios. Diario Oficial de la Unión Europea, 139 de 30.04.2004: 1-54. Recuperado de: https://www.boe.es/doue/2004/139/L00001-00054.pdf UE [Unión Europea], 2004b, Reglamento (CE) n.º 853/2004 del Parlamento Europeo y del Consejo de 29 de abril de 2004 por el que se establecen normas específicas de higiene de los alimentos de origen animal. Diario Oficial de la Unión Europea, 139 de 30.04.2004: 55-205. Recuperado de: https://www.boe.es/doue/2004/139/ L00055-00205.pdf Vilaboa-Arroniz, J., 2013, La ganadería doble propósito desde una visión agroecosistémica. Agro-productividad, 6: 9-15. Villarroel-Molina, O., Barba-Capote, C., García Martínez, A. R., & Rangel, J., 2019, Use of social networks to explore smallholder’s adoption of technologies in dual purpose farms. ESIC MARKET Economic and Business Journal, (Volume 50, Issue 2), 233–257. doi:10.7200/esicm.163.0502.1