Towards Digital Farming: Exploring Technological Integration in Agricultural Practices of a 1 sample of Italian livestock farms. 2 Ogochukwu Felicitas, Okoye *1; Selene, Righi 1; Colomba, Sermoneta 2; Gianluca, Brunori 1; Michele, 3 Moretti1. 4 1 University of Pisa/Department of Agriculture, Food and Environment, Pisa, Italy 5 2 National Institute of Statistics (ISTAT), Italy 6 * Corresponding author: Ogochukwu Felicitas Okoye (felicitas.okoye@phd.unipi.it) 7 This article has been accepted for publication and undergone full peer review but has not been 8 through the copyediting, typesetting, pagination and proofreading process, which may lead to 9 differences between this version and the Version of Record. 10 11 Please cite this article as: 12 Okoye OF, Righi S, Sermoneta C, Brunori G, Moretti M (2025). Towards Digital Farming: 13 Exploring Technological Integration in Agricultural Practices of a sample of Italian livestock 14 farms, Just Accepted. DOI: 10.36253/bae-16777 15 16 17 Abstract 18 Despite the rapid rise of digital technologies in agriculture, their application remains more 19 prominent in crop farming than in the livestock sector. Recognizing this gap, our study explores 20 the current state and determinants of digital technology adoption across Italian livestock farms, 21 examining key factors and broader trends in the industry. Using national agricultural census, and 22 national statistical programme data, we applied a logistic regression model to assess the 23 likelihood of adoption of technology. Findings reveal that large ruminant farms, particularly dairy 24 cattle and buffalo, are more likely to integrate digital tools like decision support systems, cloud 25 services, and monitoring devices. In contrast, meat cattle, small ruminants, and pig farms lag. 26 Key determinants include broadband connectivity, ownership structure, education, and age, with 27 additional factors influencing specific technology categories. Our results establish a foundation 28 for future policy and investment, underscoring the need to build digital infrastructure and promote 29 an inclusive model. 30 Keywords: Digital agriculture, Livestock farming, Technology adoption. 31 JEL codes: O33, Q16, C83. 32 33 mailto:felicitas.okoye@phd.unipi.it 1 Introduction 34 The livestock industry has undergone significant transformations driven by evolving human 35 needs, changes in consumption (Righi et al., 2023), and technological advancements (Subach & 36 Shmeleva, 2022). Traditionally, livestock farming relied on subjective and less quantifiable but 37 holistic approaches to animal welfare and management (Buller et al., 2020). However, a shift 38 towards modern, data-driven processes characterized by interconnectivity and efficiency marks 39 a significant change. Often referred to as digital farming, this shift includes concepts such as 40 smart farming, precision agriculture, and Agriculture 4.0, rooted in the sustainability discourse 41 and bolstered by advancements in information technology (Zhou et al., 2022). This shift is driven 42 by concerns over food insecurity, economic factors, climate change, market dynamics and 43 sustainability (Thornton et al., 2014). 44 Digital technologies advance agricultural practices by enabling diagnosis, intelligent perception, 45 decision-making, and improved production processes (Zhou et al., 2022; Finger, 2023). Applied 46 along the entire value chain, they enhance resource management, trade promotion, operational 47 efficiency, and knowledge exchange (Barnes et al., 2019; Elijah et al., 2018), promoting 48 sustainability, resilience, and overall competitiveness (Finger, 2023). However, adoption 49 challenges persist. Finger (2023) notes that technologies with more potential benefits are least 50 profitable and less widely adopted by small scale farmers due to their high investment costs and 51 limited returns. Large and capital-intensive farms tend to benefit disproportionately from these 52 technologies, exacerbating existing inequalities (Hackfort, 2021). Additional challenges include 53 data ownership and power distribution within the farming community (Morrone et al., 2022; 54 Neethirajan & Kemp, 2021; Wolfert et al., 2017). 55 Despite the growing integration of digital technologies in agriculture, livestock farming remains 56 one of the least digitalized sectors globally (Neethirajan & Kemp, 2021). Meanwhile, adoption 57 is diverse and limited to basic aspects of livestock management (Guntoro et al., 2019). Barriers 58 to adoption include social factors such as low levels of trust in new technologies, digital illiteracy, 59 and resistance to change (Eastwood et al., 2021; FAO, 2022), as well as technological challenges 60 like infrastructure limitations and system integration, compatibility, and interoperability (Abeni 61 et al., 2019; Tuyttens et al., 2022). Economic factors include high cost of technology, uncertainty 62 about the return on investment, and misaligned business models (Groher et al., 2020). 63 Additionally, the complexity of agroecosystems, compounded by difficulties in collaboration 64 among stakeholders, agreeing on common goals and farming practices represent barriers to 65 successful digitalization (Grivins & Kilis, 2023). These combined factors create a complex 66 environment that limits widespread adoption of digital technologies in livestock farming (Cui & 67 Wang, 2023). Moreover, digital technology development is dominated by a small number of 68 powerful multinational corporations, prioritizing digital innovations that support their market 69 dominance, potentially misaligning with the practical needs of smaller farms (Hackfort 2021). 70 Research related to digitalisation in the livestock sector has focused more on certain livestock 71 species, particularly cattle, with less emphasis on small ruminants like sheep and goats (Morrone 72 et al., 2022; Tzanidakis et al., 2023) and other livestock such as pigs and buffaloes. This disparity 73 has implications for animal welfare, leading to species- or livestock- specific differences 74 (Tuyttens et al., 2022). Therefore, it is crucial to comprehend the extent of digital technology 75 usage in livestock production and management (Fuentes et al., 2022). In Italy, the livestock sector 76 is vital not only economically but also for its cultural heritage (Pulina et al., 2017). Local breeds, 77 adapted through selective breeding, support sustainable husbandry and rural communities. By 78 leveraging breed diversity and innovative technologies, the sector can address challenges like 79 greenhouse gas emissions and resource management, contributing to both productivity and 80 sustainability. 81 The primary goal of this research is to explore the extent of digitalization among Italian livestock 82 farms and identify the factors influencing digital technology adoption. This research contributes 83 to existing literature in several ways. First, the study provides a comprehensive analysis across 84 all livestock categories, including small ruminants and monogastric species, thereby addressing 85 a critical gap in digitalization research within the livestock sector, predominantly focused on 86 dairy cattle and large ruminants. Second, by leveraging a large-scale, nationally representative 87 datasets, we offer robust empirical evidence on the determinants of digital technology adoption, 88 moving beyond the commonly acknowledged availability of digital solutions to assess the actual 89 uptake and utilization of these technologies at the farm level. Third, this study introduces a new 90 economic dimension by examining the role of marketing channels and their influence on digital 91 investments, shedding light on how different marketing sales strategies may either facilitate or 92 hinder the adoption of technological innovations. The findings of this research provide valuable 93 insights for policymakers, technology providers, and researchers seeking to foster a more 94 inclusive and effective digital transformation in livestock farming. Specifically, it seeks to 95 answer: What is the current extent of digital technology usage in livestock production and 96 management? What factors influence the adoption these technologies? The paper goes on with 97 a literature review on how digital technologies have been integrated into the livestock sector over 98 time. In section 3, the material and methods are explained. The results of the analysis are 99 presented and discussed in section 4. We conclude with reflections on policy implications and 100 recommendations. 101 2 Digitalization in the livestock sector 102 Digitalization in the livestock sector encompasses a broad array of technological innovations 103 reliant on digital infrastructures and networks. These components enhance the creation, storage, 104 and exchange of data, improving the functionality of various digital tools. Initially, Information 105 and Communication Technologies (ICT) were pivotal in precision agriculture (Cox, 2002), but 106 today’s digital solutions leverage the internet, big-data technology, cloud computing, and the 107 internet of things (IoT) through sensors and wireless communication networks (Zhou et al., 108 2022). 109 Technologies in this digital shift, as recorded in the early 2000s, involve basic data management 110 systems, including electronic identification systems such as Radio Frequency Identification 111 (RFID) tags, tracking collars, and wearable biosensors, enabling farmers to monitor individual 112 animal health, behaviours, and movement (Eastwood et al., 2021). These tools have improved 113 animal welfare and productivity through early detection of health issues (Zhou et al., 2022). In 114 Europe, Farm Management Information Systems (FMIS), barn cameras, and sensors are 115 increasingly integrated into livestock management (Gabriel & Gandorfer, 2023). This phase also 116 introduced simple automation for feeding, milking, and climate control systems. 117 Building on these early innovations, the mid-to-late 2000s saw the development of automated 118 farm management technologies, increased development of robotic systems, and the integration 119 of real-time data collection tools. These advancements laid the groundwork for further 120 digitalization, allowing farms to transition from basic monitoring to more interactive, data-driven 121 decision making. 122 In the last decade, digitalization advanced with integrated systems combining data analytics, real-123 time monitoring and advanced sensors to enhance precision and decision making (Wolfert et al., 124 2017; Chavan et al., 2024). Technologies at the forefront include Artificial Intelligence (AI)-125 driven analytics, IoT-based monitoring systems, blockchain technology, advanced automation, 126 and robotics, which promise improvement in sustainability and traceability (Calcante et al., 2014; 127 Alonso et al., 2020a; D’Agaro et al., 2021; Alshehri, 2023; Alipio & Villena, 2023). However, 128 blockchain technology is in its early stage of development and requires further validation and 129 adoption (Neethirajan & Kemp, 2021). Further innovations such as gene editing and advanced 130 biometrics offer further potential for improving animal welfare and sustainability (Eastwood et 131 al., 2021). As shown in figure 1, these digital tools address species-specific challenges and 132 benefits various livestock production systems - commodity based, nature-based, subsistence, and 133 value-seeking systems (Kraft et al., 2022) and tailored technological solutions are necessary for 134 each system as they face unique challenges (Baker et al., 2022). For instance, a commodity-based 135 system might struggle with regulatory issues, while a value-seeking system may face challenges 136 related to skill shortages and data management. Despite these advancements, concerns remain 137 that the focus on high-tech solutions may neglect simpler, more accessible innovations that could 138 benefit a wider range of farmers, particularly small-scale operations (Barrett & Rose, 2022). To 139 address issues of exclusion, Alonso et al., (2020b) suggest creating an integrated technology 140 ecosystem — a coordinated network — where IoT, edge computing, AI, and blockchain 141 technologies are bundled together. This approach is intended to lower the costs associated with 142 adopting these technologies individually. However, while such an ecosystem could enhance 143 accessibility, it may also introduce privacy and data ownership concerns and limit widespread 144 adoption. Promoting individual components of precision livestock farming technologies can be 145 unsustainable (Banhazi et al., 2012). 146 Digitalization presents both optimistic prospects and contentious challenges. Precision livestock 147 farming technologies can enhance animal welfare monitoring, but increased automation risks 148 detaching farmers from direct animal care (Buller et al., 2020; Tuyttens et al., 2022), and 149 technologies like drones may also cause stress to animals (Alanezi et al., 2022). Ethical concerns 150 persist around factory farming practices and environmental impacts (Neethirajan & Kemp, 2021). 151 Meanwhile the prioritization of economic efficiency could undermine broader sustainability 152 goals (MacPherson et al., 2022). While these tools can drive sustainability, their deployment 153 must prioritize ethical standards and resource efficiency. 154 Moreover, digitalization is reshaping farm business models, facilitating direct marketing and 155 mitigating price effects and risks. However, it also alters farm operations, work content, and 156 interactions among value chain stakeholders (Weber et al., 2022), requiring support for farmers as 157 they navigate these complexities. While the adoption of digital technologies is inevitable, 158 particularly as they become more affordable, akin to the widespread adoption of smartphones 159 (Morrone et al., 2022), farmers adoption decisions are influenced by socio-demographic, 160 institutional, and economic factors, which complicate the process (Vecchio et al., 2020). 161 The future of digital technologies in livestock farming requires a balanced approach that 162 recognizes current capabilities while learning from past lessons. Stakeholders must integrate 163 digital advancements with practical applications and ethical considerations to build a sustainable 164 and competitive future (Eastwood et al., 2021). 165 Figure 1. Livestock production systems: technologies and impacts. Adapted from Baker et al., (2022). 166 167 168 3 Material and methods 169 3.1 Data 170 This study utilized secondary data from two sources: the 7th General Census on Agriculture 171 conducted by the National Institute of Statistics (Istituto Nazionale di Statistica-ISTAT) for the 172 agricultural year 2019-2020, specifically covering the period from November 1st, 2019, to 173 October 31st, 2020. The reference date for farm herd size in this dataset was December 1st, 2020. 174 This census provided structural data on farms at national, regional, provincial, and municipal 175 levels. The ISTAT survey was structured into sections covering general farm information, land 176 use, consistency of farms, farm management methods, related activities and company managers, 177 labor, and additional information on product destination, revenues/subsidies, marketing, 178 investment in innovations, and memberships in associations or organizations. This dataset 179 contributed to most of the socio-demographic variables used for this study. The target population 180 comprised Italian farms, with at least 20 Ares (2,000 m²) of Utilized Agricultural Area (UAA). 181 The National Statistical Programme (IST-00173) survey (NSP) aimed to survey the number of 182 cattle, buffaloes, pigs and sheep and goats owned on June 1st and December 1st of each calendar 183 year. In the December 2020 edition, additional questions were introduced to survey the diffusion 184 of technologies in livestock holdings. The livestock farms are selected by stratified random 185 sampling; the stratification is by region and livestock classes, based on the total number of 186 animals owned on the farm. Both datasets were collected via online questionnaire and telephone 187 interviews. 188 The datasets initially included 246,161 observations from ISTAT and 7,587 from the NSP 189 survey. After merging based on identification number and cleaning the datasets, the sample size 190 was reduced to 4,133 livestock farms. The following exclusion criteria were further applied, that 191 is, farms raising multiple types of livestock (due to the ambiguity of digital technology 192 application), and those raising only horses, rabbits, ostriches, or poultry. The final sample size 193 was 2,412 observations. The variables used in the analysis, summarized in Table 1, were 194 categorized based on their relevance to technology adoption, as highlighted in the existing 195 literature. 196 Table 1. Variables selected from the datasets. 197 Variable Type Description SOCIO-DEMOGRAPHIC CHARACTERISTICS Age Continuous Individual age; Min. age: 21 & Max. age: 94 Gender Categorical 1: Male; 2: Female Education Categorical 1: No educational qualification. 2: Primary school leaving certificate/ Certificate of final assessment. 3: Secondary or vocational school-leaving certificate (obtained no later than the year 1965) /Secondary education diploma. 4: Professional education/training qualification related to agriculture that does not permit enrolment in university. 5: Professional education/training qualification not related to agriculture that does not permit enrolment in university. 6: Upper secondary education related to agriculture that allows enrolment in university. 7: Upper secondary education not related to agriculture that allows enrolment in university. 8: Bachelors/master’s or equivalent qualification in agriculture. 9: Bachelors/master’s or equivalent qualification not in agriculture. Area Categorical 1: Central; 2: Northeast; 3; Northwest; 4: South. FARM OPERATIONS & ENDOWNMENT Labor Continuous Total number of employees on the farm. Livestock unit (UBA) Continuous Sum of each livestock unit. Total agricultural area (SAT) Continuous Total area in hectares Type of ownership (Legal form) Categorical 1: Entrepreneur or sole proprietorship or family business. 2: Partnerships 3: Capital companies 4: Cooperative Societies 5: State Administration or Public Body. 6: Other private entities 7: Collective ownership 8: Consortia Production system Categorical Organic=1, conventional=2 Type of livestock (Dairy cattle, meat cattle, Buffaloes, Sheep, Goat, Pig) Categorical 1: large ruminants; 2: Small ruminants; 3: Monogastric Number of livestock Continuous Total number of each livestock in the farm. Membership of association Binary 0: No association; 1: Yes association Association with producer organization Association with network of enterprises Association with other organization/companies Public subsidies (%) Continuous The amount of subsidy received in percentage of the farms total gross revenues. Marketing channels for animal products Binary 0: No sales on this marketing channel. 1: Yes, sales in this marketing channel. Direct sales on the farm Off-farm direct sales Sales to other farms Sales to industrial companies in Free Market Sales to industrial companies with multi-annual agreements Sales to commercial enterprises in Free Market Sales to commercial enterprises with multi-annual agreements Sales or contribution to association bodies Other Remunerative activities Binary 0: No other activities, 1: Yes, Other activities BEHAVIOURAL ATTRIBUTE Agricultural training (Learning orientation) Binary 0: Never attended agricultural training. 1: Attended agricultural training. REGIONAL INFRASTRUCTURES Fixed broadband connectivity Binary 1: Yes, the farm uses at least one fixed broadband internet connection, 0: No DIGITALIZATION . Binary 0: No, I do not use this technology. 1: Yes, I use this technology. Decision support software Data Cloud computing Digital devices for animal monitoring Social network/website Precision animal husbandry system/machinery 198 199 3.2 Methodology 200 The study draws on various explanatory variables established in the literature to explore the 201 adoption of digital technologies in livestock farming. Tey & Brindal, (2012) classify factors 202 influencing adoption into six categories: socio-economic, agro-ecological, institutional, 203 informational, perceptual, and technological. Based on these, a tailored set of variables were 204 selected for this research. 205 The analysis categorized livestock into large ruminants, small ruminants and monogastric 206 animals because they have distinct management requirements, economic values and level of 207 environmental impacts that shape the development of digital technology. Variations in the 208 availability and adoption of digital technologies across different livestock categories have been 209 documented in the literature (Thomann et al., 2023). Research on digital technologies has 210 predominantly focused on dairy cattle reflecting their intensive management systems and 211 economic importance (Marino et al., 2023). Similarly, buffalo farming follows specialized 212 intensive management strategies, particularly in regions where it plays a significant economic 213 role, such as Italy, due to the high demand for Mozzarella di Bufala Campana PDO (Trapanese 214 et al., 2024). While digitalization efforts often focus on ruminants due to their large land use and 215 contributions to greenhouse gas (GHG) emissions (Pulina et al., 2017), this study considers 216 adoption across all livestock categories to assess variation. 217 In assessing the status of digitalization in farms, the study considered the data utilization 218 (meteorological, satellite, and drone-collected data), websites and social networks, cloud 219 computing, Decision Support Systems (DSS), digital devices for individual animal monitoring, 220 and precision husbandry systems or machinery. The percentage frequency of digital technology 221 adoption was then calculated for each livestock category. 222 To ensure robustness and comparability across livestock categories in analyzing the determinants 223 of technology adoption, the study further identified specific tools within digital devices for 224 animal monitoring and precision animal husbandry systems/machinery. These technologies were 225 analyzed as individual tools, rather than being grouped under broader categories. Only 226 technologies applicable to all livestock types were retained in the analysis. Species-specific 227 technologies, such as milking robots and milking parlors equipped with online milk quality 228 measurement systems (which are exclusively used in dairy farming), were excluded to maintain 229 consistency and enable cross-category comparisons. Similarly, social networks and websites 230 were not considered core digital tools, as they do not serve a direct farm management function. 231 Technologies were classified into three categories: highly adopted (adopted by at least 15% of 232 sampled farms), somewhat adopted (adopted by less than 15% of sampled farms) and no digital 233 technologies. Accordingly, livestock farms were grouped into farms using highly adopted digital 234 technologies, farms using somewhat adopted digital technologies, and farms using no digital 235 technology. This classification enabled a detailed assessment of digital technology adoption 236 across farms by capturing variations in uptake. Including the ‘somewhat adopted’ category 237 allowed us to distinguish farms with low but notable level of adoption, enabling a more focused 238 analysis of the factors influencing digital technology adoption. This approach helped identify 239 specific drivers or barriers associated with each category. This categorization formed the 240 dependent variable in the econometric analysis, where the outcome variable 𝑌𝑖 was binary, taking 241 a value of 1 if the farm adopted any form of digital technology (highly or somewhat adopted) 242 and 0 if no digital technology was adopted. 243 3.2.1 Model specification 244 A binary choice framework was used to model the decision to adopt digital technologies. A 245 logistic regression model was employed to estimate the probability of adoption based on a set of 246 independent variables. This approach, estimated via the maximum likelihood estimation (MLE), 247 is preferred over multinomial logistic regression, as the categorization of digital technology use 248 is not based on a distinct choice between unordered categories, but rather a binary decision of 249 adoption. The logistic regression model is specified as: 250 log ( (Pr( 𝑌𝑖=1)) 1−Pr(𝑌𝑖 =1) ) = 𝛽0 + 𝛽1𝑋1 + ⋯ + 𝛽𝑛 𝑋𝑛 (1) 251 Where: 252 Pr( 𝑌𝑖 = 1) represents the probability that farm 𝑖 adopts at least one digital technology within 253 each category. 𝑋1, … , 𝑋𝑛 are the independent variables hypothesized to influence adoption. 𝛽0 is 254 the intercept, and 𝛽𝑛 (for each 𝑛) are the estimated coefficients indicating the change in log-odds 255 with a unit change in each variable. The independent variables include, total area of holding, 256 state aid subsidy received, livestock unit (UBA), age, labour, total number of livestock, fixed 257 broadband connectivity, area, type of livestock, type of ownership (legal form), production 258 system, other remunerative activities, gender, education, agricultural training, membership of 259 association, and marketing channels. 260 3.2.2 Marginal Effects 261 Marginal effects were calculated to provide an intuitive understanding of how each variable 262 impacts the likelihood of adopting digital technologies. These effects translate the logistic 263 regression coefficients into changes in the probability of adoption. 264 The marginal effect of each variable 𝑋𝑗 on the probability of adoption 𝑃𝑟(𝑌 = 1|𝑋) is given by: 265 𝜕Pr(𝑌 = 1|𝑋) 𝜕𝑋𝑗 = 𝑃𝑟(𝑌 = 1|𝑋) . (1 − 𝑃𝑟(𝑌 = 1|𝑋)) . 𝛽𝑗 (2) 266 Where: 267 𝜕Pr(𝑌 = 1|𝑋) 𝜕𝑋𝑗 is the marginal effect of variable 𝑋𝑗 on adopting probability. 268 𝑃𝑟(𝑌 = 1|𝑋) is the predictive probability of adoption. 269 1 − 𝑃𝑟(𝑌 = 1|𝑋) is the probability of non-adoption. 270 𝛽𝑗 is the estimated coefficient of 𝑋𝑗. 271 Marginal effects were computed for both highly adopted and somewhat adopted technologies to 272 observe differential impacts across varying levels of adoption. 273 4 Results and Discussions 274 4.1 Descriptive results 275 Table 2 presents the descriptive analysis of the sampled livestock farms. The average age varied 276 across livestock types, ranging from an average of 47 years for buffalo farmers to an average of 277 56 years for pig farmers, with an overall average age of 53 years. This reflects the typical 278 demographic trend of middle-aged individuals managing farms. Gender distribution shows a 279 predominance of male farm managers (85%), consistent with traditional gender roles in 280 agriculture. 281 Labor input, measured as the number of permanent employees, also varied across livestock 282 categories, with buffalo farms requiring the highest average labor (4.6 employees), while other 283 types, such as meat cattle and sheep farms, operated with fewer workers. The scale of operations, 284 measured in livestock units, showed significant variation, with buffalo farms having the largest 285 average livestock units per farm (321.8), indicating a high scale of operations that requires more 286 labor force. There are also variations in the total area of holdings (SAT). 287 Public subsidies, an essential component of agricultural support, varied across livestock types. 288 These variations could be attributed to policy priorities or the socio-economic focus of subsidies. 289 Ownership structures leaned heavily towards sole proprietorship and family-owned businesses, 290 which likely offer flexibility in decision-making processes that are less bureaucratic than other 291 ownership forms. In terms of education, 41.3% of farmers had secondary education diplomas, 292 and only 17% had formal higher education in agriculture. However, 64.9% of farmers 293 participated in agricultural training programs, indicating a reliance on non-formal education to 294 compensate for the lower levels of formal education. 295 Conventional farming remained the most predominant type of production system (95.2%), while 296 organic farming remained relatively limited (5%). The limited adoption of organic methods may 297 be due to the challenges of organic livestock production, as farmers rely on organic crops and 298 avoid conventional feed additives, hormones, and medicines, making organic farming both costly 299 and complex. Engagement in other remunerative activities varied by livestock type, with goat 300 farmers (34.52%) participating more frequently in these activities compared to other livestock 301 types. A suggestion that income diversification may be a crucial strategy for economic resilience. 302 Marketing channels for final products were influenced by farm scale and final products 303 (processed vs. non-processed), and 45.4% of respondents were affiliated with companies or 304 organizations, with 28.2% participating in producer associations, reflecting a differentiated 305 approach to market engagement. 306 Table 2. Descriptive statistics of socio-economic characteristics across various livestock types. 307 Dairy Cattle (n=1091) Meat Cattle (n=435) Buffalo (n=49) Goat (n=84) Sheep (n=329) Pig (n=424) Total (n=2412) Age (mean & SD) 54 (13.4) 53 (14.2) 47 (13.2) 49.6 (15.5) 48.6 (14.4) 55.7 (14.0) 53.0 (14.0) Gender Male 987 355 43 62 266 350 2063 Female 104 80 6 22 63 74 349 Labour (mean & SD) 3.2 (3.7) 1.6 (2.4) 4.6 (3.7) 1.6 (2.1) 1.9 (5.6) 2.5 (5.5) 2.6 (4.2) Type of ownership (legal form) Entrepreneur or sole proprietorship or family business 525 356 30 70 259 273 1513 Partnerships 543 69 12 11 63 132 830 Capital companies 17 7 7 2 4 16 53 Cooperative Societies 6 3 - 1 3 3 16 State Administration or Public Body - - - - - - - Other private entities - - - - - - - Collective ownership - - - - - - - Consortia - - - - - - - Education No educational qualification 4 5 - - 5 4 18 Primary school leaving certificate/ Certificate of final assessment 90 67 - 9 35 65 266 Secondary education diploma 458 183 16 39 145 155 996 Professional education/training qualification related to agriculture 77 14 - 1 9 14 115 Professional education/training qualification not related to agriculture 57 18 2 3 10 26 116 Upper secondary education related to agriculture 156 44 6 13 31 45 295 Upper secondary education not related to agriculture 167 63 19 11 73 79 412 Bachelor’s/master’s or equivalent qualification in agriculture 41 12 1 1 9 15 79 Bachelor’s/master’s or equivalent qualification not in agriculture 41 29 5 7 12 21 115 Livestock unit (UBA) (mean & SD) 196.0 (231.9) 81.4 (247.6) 321.8 (250.9) 20.1 (30.8) 44.6 (50.5) 248.9 (697.2) 160.4 (358.3) Production system Organic 33 45 4 4 19 10 115 Conventional 1058 390 45 80 310 414 2297 Total No of livestock (mean & SD) 128.5 (155.2) 66.0 (295.7) 342.6 (262.0) 115.1 (194.3) 398.8 (480.2) 868.9 (2236.2) 288.2 (1010.2) Total area of holding (SAT) ha. (mean & SD) 68.9 (80.1) 84.9 (169.8) 45.3 (50.6) 35.3 (52.4) 81.6 (160.2) 50.4 (229.8) 68.7 (145.5) Membership of association Producer association 398 99 16 19 54 93 679 Network of enterprises 45 11 5 3 12 15 91 Association with companies/organization 579 160 22 23 130 181 1095 Public subsidies (%) 13.2 (17.3) 225.1 (24.5) 8.0 (12.0) 18.0 (26.9) 30.2 (28.8) 13.3 (20.4) 17.7 (22.4) Marketing channels Direct sales on the farm 135 776 10 18 60 112 411 Off-farm direct sales 59 32 4 4 31 43 173 Sales to other farms 65 78 1 10 19 32 205 Sales to industrial companies in Free Market 220 39 14 11 69 59 412 Sales to industrial companies with multi- annual agreements 75 11 1 2 9 9 107 Sales to commercial enterprises in Free Market 328 179 12 14 91 91 715 Sales to commercial enterprises with multi- annual agreements 33 7 3 1 0 3 47 Sales or contribution to association bodies 398 31 6 8 48 17 508 Remunerative activities 215 62 8 29 44 132 490 Agricultural Trainings (Learning orientation) 827 231 36 40 173 258 1565 Regional Infrastructures Fixed broadband connectivity (%) 64.5 37.5 63.3 45.2 28.9 51.2 51.7 Source: own elaboration. 308 4.2 Extent of digital technology adoption 309 Table 3 illustrates significant variation in digital technology adoption across livestock types, 310 revealing distinct trends within the Italian livestock sector. Decision Support Systems (DSS) are 311 adopted by 28.32% of farms, indicating their role in optimizing farm management and decision-312 making. Similarly, cloud computing services are utilized by 26.16% of farms, with the highest 313 adoption rates observed among buffalo (48.98%) and dairy cattle farms (33.09%), emphasizing 314 their role in managing complex, large-scale operations. In contrast, meat cattle, sheep, goat, and 315 pig farms display relatively lower adoption rates. The limited uptake in these sectors may reflect 316 a low perceived value of investment in digital technologies, as these farms may not require the 317 same level of operational complexity as dairy or buffalo farms. 318 However, pig farms (24.06%) report a higher adoption of cloud services compared to meat cattle 319 (17.47%), suggesting a moderate level of digitalization in this sector. Among small ruminants, 320 goat farms (23.81%) exhibit a higher adoption rate of cloud services compared to sheep farms 321 (14.59%). 322 In the sample, 29.98% of the farms use any type of digital devices, particularly sensors for 323 individual animal monitoring that is widely adopted among dairy (37.21%) and buffalo farms 324 (28.57%), where real-time tracking of animal production and health is essential in high-output 325 dairy operations. In contrast, meat cattle (20.69%), sheep (19.15%), and goat farms (17.86%) 326 show moderate adoption, while pig farms report the lowest uptake (9.91%). Detectors for 327 individual animal production are primarily employed in buffalo (16.33%) and dairy cattle farms 328 (15.12%), whereas behavior image analyzers show minimal adoption across all species (3.69%). 329 Precision animal husbandry systems remain underutilized, with an overall adoption rate of 330 1.78%. Their adoption is virtually nonexistent among small ruminants, with no recorded uptake 331 in goat and sheep farms. Among large ruminants, dairy cattle (2.84%) and buffalo farms (2.04%) 332 report some adoption, while meat cattle and pig farms have minimal use (1.38% and 1.18%, 333 respectively). According to Bucci et al., (2019), Italy continues to lag behind other EU countries 334 in the adoption of Precision Agriculture Technologies (PATS). Additional precision systems, 335 such as information management tools (0.95%), machines with online food quality analysis 336 (0.12%), remote animal identification (0.54%), and robotic systems for ration management and 337 stable cleaning (0.21%), remain rare across all livestock categories. The limited use of Precision 338 Livestock Farming (PLF) technologies in small ruminants and pig farms suggests significant 339 untapped potential for digital transformation in these sectors. Studies by Abeni et al., (2019) and 340 Vaintrub et al., (2021) indicate that advanced management tools are predominantly adopted in 341 dairy-intensive regions, where they enhance operational efficiency and reduce labor costs, 342 although this trend remains less prevalent in the Italian livestock sector. 343 The adoption of meteorological and satellite data follows a similar pattern, with dairy cattle farms 344 (26.58%) being the primary users, followed by meat cattle (18.39%) and buffalo farms (14.29%). 345 Small ruminant and pig farms demonstrate relatively low adoption rates, indicating a reduced 346 reliance on environmental data for operational decision-making. Large, commercially oriented 347 farms tend to perceive greater value in meteorological data, whereas smaller-scale and traditional 348 operations may not prioritize these digital tools. 349 Social networks and websites, which are frequently used for marketing and supply chain 350 engagement, show the highest adoption rates among goat (27.38%) and buffalo farms (26.53%). 351 In contrast, dairy cattle (12.37%), meat cattle (7.36%), sheep (8.51%), and pig farms (19.34%) 352 exhibit lower engagement on these platforms, which may reflect sector-specific market structures 353 and differences in digital literacy. 354 In summary, the results reveal clear disparities in digital technology adoption across livestock 355 sectors. Dairy cattle and buffalo farms lead in the adoption of DSS (43.26% and 44.90%, 356 respectively) and cloud computing (33.09% and 48.98%, respectively), indicating higher levels 357 of digital transformation in these intensive production systems. Conversely, meat cattle, small 358 ruminants, and pig farms demonstrate limited digital engagement, highlighting areas of untapped 359 potential and the need for targeted strategies to promote digital technology adoption. 360 Table 3. Adoption of digital tools across different livestock types in percentages. 361 Digital tools Dairy Cattle (n=1091) Meat Cattle (n=435) Buffalo (n=49) Goat (n=84) Sheep (n=329) Pig (n=424) Total (n=2412) Decision support system 43.26 16.55 44.90 11.90 9.12 18.16 28.32 Use of Data 26.58 18.39 14.29 11.90 12.46 13.44 20.11 Cloud services 33.09 17.47 48.98 23.81 14.59 24.06 26.16 Social network/website 12.37 7.36 26.53 27.38 8.51 19.34 12.98 Digital devices 41.61 23.45 40.82 17.86 23.40 12.97 29.98 Specific digital devices common to various livestock types. Sensors on the limbs neck or ear tags 37.21 20.69 28.57 17.86 19.15 9.91 26.12 Detectors of individual production 15.12 1.84 16.33 2.38 0.61 1.89 8.00 Behavior image analyzers 7.15 0.92 4.08 0.00 0.91 0.47 3.69 Others 1.01 2.76 6.12 0.00 8.21 0.71 2.32 Precision animal husbandry system 2.84 1.38 2.04 0.00 0.00 1.18 1.78 Specific precision animal husbandry system/machinery common to various Livestock types Information system for livestock management 1.92 0.46 0.00 0.00 0.00 0.00 0.95 Machines equipped with online analysis of food quality 0.18 0.23 0.00 0.00 0.00 0.00 0.12 Milking robots 0.37 0.00 2.04 0.00 0.00 0.00 0.21 Milking parlor equipped with online milk quality measurement system 0.73 0.00 2.04 0.00 0.00 0.00 0.37 Remote animal identification management 0.92 0.23 2.04 0.00 0.00 0.24 0.54 systems Sensors for detecting the productive and reproductive activity of the live- 1.83 0.46 0.00 0.00 0.00 0.24 0.95 Stock Robotic systems for ration management and stable cleaning 0.09 0.23 0.00 0.00 0.00 0.71 0.21 Others 0.18 0.00 0.00 0.00 0.00 0.71 0.21 Source. Authors own elaboration 362 4.3 Determinants of digital technology adoption in livestock production 363 The logistic regression model was used to examine factors influencing the likelihood of adopting 364 digital innovations among livestock farms. It explicitly distinguishes between farms with highly 365 adopted and somewhat adopted digital innovations. The output of the marginal effects analysis 366 is summarized in Table 4. 367 Fixed broadband connectivity significantly impacts both highly and somewhat adopted digital 368 innovations. Sozzi et al. (2021) agree that poor internet coverage impacts adopting digital 369 agriculture (DA) tools, particularly in Italy’s marginal and hilly agricultural areas. Thus, targeted 370 investments in rural internet infrastructure are critical to advancing agricultural modernization. 371 The variable Area captures regional effects on digital innovation adoption, and it is significant 372 only for highly adopted digital innovations. Compared to the Central region, the Northeast region 373 and Northwest region have a 9% and 8% likelihood of adoption of digital technology, 374 respectively. This regional disparity mirrors findings by Altamore et al., (2024), who suggest that 375 policy support and socio-economic differences shape diverse regional practices. Just as structural 376 characteristics such as governance and historical influences drive varying agricultural practices 377 across northern and southern Italy, similar regional factors may influence digital innovation 378 adoption in our context. This suggests that adoption patterns may be shaped by present-day 379 policies and longstanding regional characteristics that impact economic and social development. 380 Similarly, for the highly adopted digital innovations, the Type of livestock variable shows that 381 farmers managing monogastric animals are less likely (by approximately 14%) to adopt digital 382 technologies. Additionally, managing small ruminants is associated with a reduced likelihood 383 (by 8%) of digital technologies adoption compared to the large ruminant animals. The 384 implication is that the digital technologies available may be more applicable to certain livestock 385 systems than others and reflect specific technological needs or economic considerations in 386 different livestock types. The type of ownership (legal form) positively influences high and 387 somewhat adopted digital innovations, with a probability of 10% and 5% respectively, 388 particularly within family-owned enterprises, where streamlined decision-making processes 389 support the adoption of new technologies. Meanwhile, for somewhat adopted digital innovations, 390 Other Remunerative Activities significantly increase the likelihood of adoption by 14%, 391 suggesting that farmers with additional income streams may be more inclined to explore these 392 digital solutions, as diversified income streams provide additional financial flexibility. However, 393 these activities have no significant effect on highly adopted innovations, indicating that diversified 394 income streams do not necessarily impact the adoption of broader farm management systems like 395 DSS or cloud services. This provides evidence that although technology adoption is widespread, 396 it is often limited to basic aspects of livestock management, with more sophisticated systems 397 concentrated in specialized sectors. Education is a significant predictor for high and somewhat 398 adopted digital innovations, with each additional year of schooling increasing the likelihood of 399 adopting digital innovations by 1% in each case. Educated farmers are expected to be more 400 receptive to technological advancements, given potential improvements in understanding and 401 implementing new digital tools. Agricultural training positively impacts highly adopted digital 402 innovations (5% increase), highlighting the role that Agricultural Knowledge and Innovation 403 Systems (AKIS) play in facilitating digital transformation. In Italy, the coordination of 404 knowledge exchange and training initiatives falls under the jurisdiction of the regional authorities 405 (Birke et al., 2022). This decentralized approach allows AKIS to tailor training and knowledge-406 sharing programs for different regions' specific needs and contexts. Such regionalized systems 407 align with the Common Agricultural Policy’s cross cutting objective of modernizing agriculture 408 by fostering knowledge sharing, innovation and digitalization through specialized training 409 programs (An, 2024). State aid subsidies received show a negative impact on somewhat adopted 410 digital innovations. Altamore et al., (2024) argues that Italian farms primarily use subsidies to 411 supplement income. For instance, in southern regions of Italy where subsidies makeup a 412 substantial portion of farm revenue, farms may prioritize essential operational costs over 413 technology investment. While some EU aid targets agricultural innovation, the lack of subsidies 414 specifically for digital technology suggests that generalized aid does not effectively drive digital 415 adoption, underscoring a need for targeted incentives. Being part of a Producer association is 416 positively associated with highly adopted digital innovations, increasing the likelihood of 417 adoption by 6%. This supports the notion that community networks facilitate knowledge 418 exchange and collective learning, making innovation adoption easy. A structured network 419 encourages technology use more than independent farms by fostering a focus on product quality 420 and accountability (Chen et al., 2021). The amount of Livestock unit (UBA) has a strong positive 421 effect on highly adopted digital innovations, particularly for large farms where technologies such 422 as DSS and cloud services offer benefits for streamlining operations and managing complex farm 423 activities. This effect aligns with findings in literature, where herd size is often linked to greater 424 investment in management tools that support efficiency at scale (Abeni et al., 2019). Age has an 425 inverse relationship with highly and somewhat adopted digital innovations. Younger farmers are 426 generally more likely to adopt digital innovations, possibly because they tend to have higher 427 digital literacy and a longer-term outlook, giving them more time to benefit from these 428 technologies. This trend is consistent with findings in the technology adoption literature (Barnes 429 et al., 2019; Michels et al., 2020). Labor shows a positive impact on somewhat adopted digital 430 innovations, with each additional worker associated with a 2% increase in the likelihood of 431 adopting these technologies like precision animal husbandry systems and digital devices (e.g., behavior 432 analyzers and production detectors, etc.), which require continuous monitoring and data interpretation. 433 Conversely, highly adopted digital technologies like DSS and cloud services, designed to automate 434 tasks and reduce labor dependency post-implementation, may appeal more to farms with fewer 435 employees. Marketing channels play a critical role in shaping digital technology adoption. Direct 436 on-farm sales have been found to positively influence somewhat adopted digital innovations, 437 while sales through association bodies are more strongly linked to highly adopted innovations. 438 However, the structure of the supply chain affects the extent to which producers benefit from 439 price fluctuations and market stability, which in turn influences their ability to invest in digital 440 technologies. Empirical evidence suggests that farmers represent a weak link in the value chain, 441 meaning that increases in retail market prices often do not translate into significant financial 442 benefits for producers (Goodwin et al., 2024). This price asymmetry may limit farmers' financial 443 capacity to adopt digital tools, particularly when market uncertainty discourages long-term 444 investments. 445 Table 4. Determinants of digital innovation adoption in the livestock sector: Marginal effects analysis. 446 Variables Marginal effects (Highly adopted digital innovations) Marginal effects (Some- what adopted digital innovations) Fixed broadband connectivity 0.16*** (0.02) 0.14*** (0.02) Area Northeast 0.09** (0.04) 0.02 (0.03) Northwest 0.08* (0.04) -0.01(0.03) South -0.01(0.04) -0.04 (0.03) Type of livestock Monogastric -0.14*** (0.03) -0.01 (0.02) Small ruminants -0.08** (0.04) 0.03 (0.03) Type of ownership (Legal form) 0.10*** (0.02) 0.05*** (0.02) Total area of holding (ha) -0.00 (0.00) -0.00 (0.00) Production system Conventional -0.05 (0.05) -0.06(0.04) Other Remunerative activities 0.01 (0.03) 0.14*** (0.02) Gender Female 0.03 (0.03) -0.00 (0.02) Education 0.01** (0.01) 0.01*** (0.00) Agricultural training 0.05* (0.02) 0.03 (0.02) State aid Subsidy received -0.00 (0.00) -0.00* (0.00) Membership of association Producer association 0.06** (0.03) 0.01(0.02) Network of enterprises 0.09 (0.07) 0.02(0.04) Association with companies/organization -0.00 (0.02) 0.02(0.02) Livestock unit (UBA) 0.00*** (0.00) -0.00 (0.00) Age -0.00*** (0.00) -0.00** (0.00) Labor 0.00 (0.00) 0.02*** (0.00) Total No. of livestock -2.69e-06 (0.00) -9.57e-06 (0.00) Marketing Channels -Direct sales on the farm -0.02 (0.03) 0.04* (0.02) -Off-farm direct sales 0.00 (0.04) 0.01(0.03) -Sales to other farms -0.02 (0.04) 0.00 (0.03) -Sales to industrial companies in Free Market 0.05 (0.03) -0.03 (0.02) -Sales to industrial companies with multi-annual agreements 0.04 (0.06) 0.03(0.04) -Sales to commercial enterprises in Free Market 0.03 (0.03) -0.01(0.02) -Sales to commercial enterprises with multi-annual agreements -0.07 (0.08) 0.06 (0.05) -Sale or contribution to association bodies 0.09***(0.03) 0.02 (0.02) Source: own elaboration. Standard error values are inside parentheses, *p<0.10, **p<0.05, ***p<0.01 447 5 Conclusion 448 This study highlights the variations in the adoption of digital technologies in livestock farming 449 across Italy, with dairy and buffalo farms leading in digital engagement, while meat cattle, 450 sheep, goat, and pig farms are lagging. This disparity in digital adoption is shaped by differences 451 in operational scale and economic structure, which collectively impact animal welfare across 452 species. Key socio-economic determinants, such as broadband connectivity, farm ownership 453 structure, education, and age, significantly influence digital technology. Younger farmers and 454 those with higher education levels are generally more inclined toward technology adoption, 455 suggesting that digital literacy and training could potentially bridge adoption gaps for other farm 456 demographics. Additionally, fostering a bottom-up, participatory co-design approach in the 457 development of digital tools can play a crucial role in aligning technology with farmers' specific 458 needs and contexts. Engaging farmers actively in the co-design process allows for the tailoring 459 of digital tools to address practical challenges, usability concerns, and operational requirements 460 from the perspective of end-users. This approach not only enhances relevance but also promotes 461 a sense of ownership and empowerment among farmers. By incorporating feedback from 462 farmers into the design of digital tools, technology providers and developers can ensure that 463 these innovations are accessible, adaptable, and directly responsive to the needs of those who 464 will implement them in the field. 465 Highly adopted digital innovations are influenced by the location, livestock type, participation 466 in agricultural training, livestock unit size, membership in producer associations, and 467 engagement in sale or contributions to association bodies. In contrast, somewhat adopted 468 innovations are driven by engagement in other income-generating activities, state aid subsidies 469 received, labor availability, and involvement in direct sales marketing. Improving broadband 470 infrastructure remains crucial to fostering widespread digital adoption, and targeting younger, 471 digitally literate farmers is key. Integrating digital literacy into agricultural training programs 472 can further ensure broad and effective engagement. The gap between traditional agricultural 473 education and the skills needed for digital farming highlights the need for modernized curricula 474 to align with evolving demands of the industry. 475 The Common Agricultural Policy (CAP) presents an opportunity to promote digital 476 transformation by providing targeted funding for rural broadband infrastructure and digital 477 literacy training. Future CAP reforms could focus on promoting digital technologies that 478 enhance sustainability and efficiency, with a particular focus on small and mid-sized farms. 479 Addressing disparities in access is essential to ensure that the benefits of digital agriculture 480 extend to farms of all sizes and types. 481 These economic constraints with regards to the market structure reinforce the need for targeted 482 policies that support both digital innovation and income stabilization measures. Ensuring that 483 farmers have more predictable market conditions and access to stable contracts may enhance 484 their ability to invest in digital technologies, fostering greater digital transformation in the 485 livestock sector. 486 Looking ahead, further research should explore the long-term effects of digital technologies on 487 farm productivity, economic outcomes, and sustainability. Additionally, it is important to 488 examine the specific barriers faced by small and medium-sized farms in adopting digital tools 489 and assess how policies such as the CAP can promote digital inclusion across the sector. 490 Understanding the adaptation of digital solutions to different farm sizes and livestock species is 491 essential for advancing precision livestock farming and ensuring technological advancements 492 benefit the entire sector, not just a select few. 493 Broader implications for the livestock sector were also noted, particularly in underrepresented 494 areas such as PLF for small ruminants and pigs. Reducing barriers to adoption in these areas is 495 critical for fully realizing the potential of digital technologies across all livestock types. 496 Collaborative efforts among technology providers, agricultural organizations, and policymakers 497 are essential to fostering a sustainable and equitable transformation in livestock farming, 498 benefiting a diverse range of farm types and sizes. 499 In summary, this study provides valuable insights into the current state of digital adoption in 500 Italian livestock farming, yet limitations such as unaccounted regional differences that may 501 affect the generalizability of the findings and the dataset, which was collected up until 2021, 502 needs to be updated to reflect more recent developments, suggest avenues for ongoing research. 503 Bridging the digital divide in livestock farming will require targeted, well-informed strategies 504 to achieve a comprehensive digital transformation across the sector, equipping farmers with the 505 tools needed to thrive in an increasingly digital agricultural landscape. 506 References 507 Abeni, F., Petrera, F., & Galli, A. (2019). A Survey of Italian Dairy Farmers’ Propensity for 508 Precision Livestock Farming Tools. Animals, 9(5), 202. https://doi.org/10.3390/ani9050202 509 Alanezi, M. A., Shahriar, M. S., Hasan, M. B., Ahmed, S., Yusuf, A., & Bouchekara, H. R. 510 (2022). Livestock management with unmanned aerial vehicles: A review. IEEE Access, 511 10, 45001–45028. 512 Alipio, M., & Villena, M. L. (2023). Intelligent wearable devices and biosensors for monitoring 513 cattle health conditions: A review and classification. Smart Health. Scopus. 514 https://doi.org/10.1016/j.smhl.2022.100369 515 Alonso, R. S., Sittón-Candanedo, I., García, Ó., Prieto, J., & Rodríguez-González, S. (2020a). 516 An intelligent Edge-IoT platform for monitoring livestock and crops in a dairy farming 517 scenario. Ad Hoc Networks. Scopus. https://doi.org/10.1016/j.adhoc.2019.102047 518 Alonso, R. S., Sittón-Candanedo, I., García, Ó., Prieto, J., & Rodríguez-González, S. (2020b). 519 An intelligent Edge-IoT platform for monitoring livestock and crops in a dairy farming 520 scenario. Ad Hoc Networks, 98, 102047. https://doi.org/10.1016/j.adhoc.2019.102047 521 Alshehri, D. M. (2023). Blockchain-assisted internet of things framework in smart livestock 522 farming. Internet of Things (Netherlands). Scopus. 523 https://doi.org/10.1016/j.iot.2023.100739 524 Altamore, L., Chinnici, P., Bacarella, S., Chironi, S., & Ingrassia, M. (2024). Current 525 Framework of Italian Agriculture and Changes between the 2010 and 2020 Censuses. 526 Agriculture, 14(9), 1603. 527 An, I. (2024). Meeting the European Union’s digital agriculture requirements. 528 Baker, D., Jackson, E. L., & Cook, S. (2022). Perspectives of digital agriculture in diverse types 529 of livestock supply chain systems. Making sense of uses and benefits. Frontiers in 530 Veterinary Science. Scopus. https://doi.org/10.3389/fvets.2022.992882 531 Banhazi, T. M., Lehr, H., Black, J. L., Crabtree, H., Schofield, P., Tscharke, M., & Berckmans, 532 D. (2012). Precision Livestock Farming: An international review of scientific and 533 commercial aspects. International Journal of Agricultural and Biological Engineering. 534 Scopus. https://doi.org/10.3965/j.ijabe.20120503.00? 535 Barnes, A. P., Soto, I., Eory, V., Beck, B., Balafoutis, A., Sánchez, B., Vangeyte, J., Fountas, 536 S., Van Der Wal, T., & Gómez-Barbero, M. (2019). Exploring the adoption of precision 537 agricultural technologies: A cross regional study of EU farmers. Land Use Policy, 80, 538 163–174. https://doi.org/10.1016/j.landusepol.2018.10.004 539 Barrett, H., & Rose, D. C. (2022). Perceptions of the Fourth Agricultural Revolution: What’s 540 In, What’s Out, and What Consequences are Anticipated? Sociologia Ruralis, 62(2), 541 162–189. https://doi.org/10.1111/soru.12324 542 Birke, F. M., Bae, S., Schober, A., Wolf, S., Gerster-Bentaya, M., & Knierim, A. (2022). AKIS 543 in European countries: Cross analysis of AKIS country. 544 Bucci, G., Bentivoglio, D., Finco, A., & Belletti, M. (2019). Exploring the impact of innovation 545 adoption in agriculture: How and where Precision Agriculture Technologies can be 546 suitable for the Italian farm system? 275(1), 012004. 547 Buller, H., Blokhuis, H., Lokhorst, K., Silberberg, M., & Veissier, I. (2020). Animal welfare 548 management in a digital world. Animals. https://doi.org/10.3390/ani10101779 549 Calcante, A., Tangorra, F. M., Marchesi, G., & Lazzari, M. (2014). A GPS/GSM based birth 550 alarm system for grazing cows. Computers and Electronics in Agriculture. 551 https://doi.org/10.1016/j.compag.2013.11.006 552 Chavan, M., Dhage, S., Gaikwad, U., Deokar, D., Lokhande, A., & Kamble, D. (2024). Digital 553 livestock farming: A review. International Journal of Advanced Biochemistry Research, 554 8(4S), 469–478. https://doi.org/10.33545/26174693.2024.v8.i4sf.1027 555 Chen, X., Ou, X., Dong, X., Yang, H., Ubaldo, C., & Yue, X.-G. (2021). Impact of farmer 556 organization forms on agricultural product quality from the perspective of technology 557 adoption. 92–99. 558 Cox, S. (2002). Information technology: The global key to precision agriculture and 559 sustainability. Computers and Electronics in Agriculture, 36(2–3), 93–111. 560 https://doi.org/10.1016/s0168-1699(02)00095-9 561 Cui, L., & Wang, W. (2023). Factors Affecting the Adoption of Digital Technology by Farmers 562 in China: A Systematic Literature Review. Sustainability, 15(20), 14824. 563 D’Agaro, E., Rosa, F., & Akentieva, N. P. (2021). New Technology Tools and Life Cycle 564 Analysis (LCA) Applied to a Sustainable Livestock Production. Eurobiotech Journal. 565 Scopus. https://doi.org/10.2478/ebtj-2021-0022 566 Eastwood, C. R., Edwards, J. P., & Turner, J. A. (2021). Review: Anticipating alternative 567 trajectories for responsible Agriculture 4.0 innovation in livestock systems. Animal. 568 Scopus. https://doi.org/10.1016/j.animal.2021.100296 569 Elijah, O., Rahman, T. A., Orikumhi, I., Leow, C. Y., & Hindia, M. N. (2018). An overview of 570 Internet of Things (IoT) and data analytics in agriculture: Benefits and challenges. IEEE 571 Internet of Things Journal, 5(5), 3758–3773. 572 Finger, R. (2023). Digital innovations for sustainable and resilient agricultural systems. 573 European Review of Agricultural Economics, 50(4), 1277–1309. 574 Fuentes, S., Gonzalez Viejo, C., Tongson, E., & Dunshea, F. R. (2022). The livestock farming 575 digital transformation: Implementation of new and emerging technologies using 576 artificial intelligence. Animal Health Research Reviews, 23(1), 59–71. 577 https://doi.org/10.1017/S1466252321000177 578 Gabriel, A., & Gandorfer, M. (2023). Adoption of digital technologies in agriculture—An 579 inventory in a European small-scale farming region. Precision Agriculture. Scopus. 580 https://doi.org/10.1007/s11119-022-09931-1 581 Goodwin, B. K., Rivieccio, G., De Luca, G., & Capitanio, F. (2024). Computing impulse 582 response functions from a copula-based vector autoregressive model: Evidence from the 583 italian agri-food value chain. Quality & Quantity, 58(2), 1779–1797. 584 https://doi.org/10.1007/s11135-023-01720-w 585 Grivins, M., & Kilis, E. (2023). Engaging with barriers hampering uptake of digital tools. 586 Italian Review of Agricultural Economics, 78(2), 29–38. 587 Groher, T., Heitkämper, K., & Umstätter, C. (2020). Digital technology adoption in livestock 588 https://doi.org/10.1007/s11119-022-09931-1 https://doi.org/10.1007/s11135-023-01720-w production with a special focus on ruminant farming. Animal, 14(11), 2404–2413. 589 https://doi.org/10.1017/S1751731120001391 590 Guntoro, B., Hoang, Q. N., & A’Yun, A. Q. (2019). Dynamic Responses of Livestock Farmers 591 to Smart Farming. IOP Conference Series: Earth and Environmental Science. Scopus. 592 https://doi.org/10.1088/1755-1315/372/1/012042 593 Hackfort, S. (2021). Patterns of inequalities in digital agriculture: A systematic literature 594 review. Sustainability (Switzerland). https://doi.org/10.3390/su132212345 595 Kraft, M., Bernhardt, H., Brunsch, R., Büscher, W., Colangelo, E., Graf, H., Marquering, J., 596 Tapken, H., Toppel, K., Westerkamp, C., & Ziron, M. (2022). Can Livestock Farming 597 Benefit from Industry 4.0 Technology? Evidence from Recent Study. Applied Sciences 598 (Switzerland). Scopus. https://doi.org/10.3390/app122412844 599 MacPherson, J., Voglhuber-Slavinsky, A., Olbrisch, M., Schöbel, P., Dönitz, E., Mouratiadou, 600 I., & Helming, K. (2022). Future agricultural systems and the role of digitalization for 601 achieving sustainability goals. A review. Agronomy for Sustainable Development, 602 42(4), 70. 603 Marino, R., Petrera, F., & Abeni, F. (2023). Scientific Productions on Precision Livestock 604 Farming: An Overview of the Evolution and Current State of Research Based on a 605 Bibliometric Analysis. Animals, 13(14), 2280. https://doi.org/10.3390/ani13142280. 606 Michels, M., Fecke, W., Feil, J.-H., Musshoff, O., Pigisch, J., & Krone, S. (2020). Smartphone 607 adoption and use in agriculture: Empirical evidence from Germany. Precision 608 Agriculture, 21(2), 403–425. https://doi.org/10.1007/s11119-019-09675-5 609 Morrone, S., Dimauro, C., Gambella, F., & Cappai, M. G. (2022). Industry 4.0 and Precision 610 Livestock Farming (PLF): An up to Date Overview across Animal Productions. 611 Sensors, 22(12), 4319. https://doi.org/10.3390/s22124319 612 Neethirajan, S., & Kemp, B. (2021). Digital Livestock Farming. Sensing and Bio-Sensing 613 Research. Scopus. https://doi.org/10.1016/j.sbsr.2021.100408 614 Pulina, G., Francesconi, A. H. D., Stefanon, B., Sevi, A., Calamari, L., Lacetera, N., Dell’Orto, 615 https://doi.org/10.3390/ani13142280 V., Pilla, F., Ajmone Marsan, P., Mele, M., Rossi, F., Bertoni, G., Crovetto, G. M., & 616 Ronchi, B. (2017). Sustainable ruminant production to help feed the planet. Italian 617 Journal of Animal Science, 16(1), 140–171. 618 https://doi.org/10.1080/1828051x.2016.1260500 619 Righi, S., Viganò, E., & Panzone, L. (2023). Consumer concerns over food insecurity drive 620 reduction in the carbon footprint of food consumption. Sustainable Production and 621 Consumption, 39, 451–465. https://doi.org/10.1016/j.spc.2023.05.027 622 Sozzi, M., Kayad, A., Ferrari, G., Zanchin, A., Grigolato, S., & Marinello, F. (2021). 623 Connectivity in rural areas: A case study on internet connection in the Italian 624 agricultural areas. 466–470. 625 Subach, T. I., & Shmeleva, Z. N. (2022). Introduction of digital innovations in livestock 626 farming. IOP Conference Series: Earth and Environmental Science, 1112(1), 012079. 627 https://doi.org/10.1088/1755-1315/1112/1/012079 628 Tey, Y. S., & Brindal, M. (2012). Factors influencing the adoption of precision agricultural 629 technologies: A review for policy implications. Precision Agriculture, 13(6), 713–730. 630 https://doi.org/10.1007/s11119-012-9273-6. 631 Thomann, B., Würbel, H., Kuntzer, T., Umstätter, C., Wechsler, B., Meylan, M., & Schüpbach-Regula, 632 G. (2023). Development of a data-driven method for assessing health and welfare in the most 633 common livestock species in Switzerland: The Smart Animal Health project. Frontiers in 634 Veterinary Science, 10, 1125806. https://doi.org/10.3389/fvets.2023.1125806. 635 Trapanese, L., Petrocchi Jasinski, F., Bifulco, G., Pasquino, N., Bernabucci, U., & Salzano, A. (2024). 636 Buffalo welfare: A literature review from 1992 to 2023 with a text mining and topic analysis 637 approach. Italian Journal of Animal Science, 23(1), 570–584. 638 https://doi.org/10.1080/1828051X.2024.2333813 639 FAO. (2022), The State of Food and Agriculture 2022, FAO. https://doi.org/10.4060/cb9479en 640 Thornton, P. K., Ericksen, P. J., Herrero, M., & Challinor, A. J. (2014). Climate variability and 641 vulnerability to climate change: A review. Global Change Biology, 20(11), 3313–3328. 642 Tuyttens, F. A. M., Molento, C. F. M., & Benaissa, S. (2022). Twelve Threats of Precision 643 https://doi.org/10.1007/s11119-012-9273-6 https://doi.org/10.3389/fvets.2023.1125806 Livestock Farming (PLF) for Animal Welfare. Frontiers in Veterinary Science. Scopus. 644 https://doi.org/10.3389/fvets.2022.889623 645 Vaintrub, M. O., Levit, H., Chincarini, M., Fusaro, I., Giammarco, M., & Vignola, G. (2021). 646 Precision livestock farming, automats and new technologies: Possible applications in 647 extensive dairy sheep farming. Animal, 15(3), 100143. 648 Vecchio, Y., De Rosa, M., Adinolfi, F., Bartoli, L., & Masi, M. (2020). Adoption of precision 649 farming tools: A context-related analysis. Land Use Policy. Scopus. 650 https://doi.org/10.1016/j.landusepol.2020.104481 651 Weber, R., Braun, J., & Frank, M. (2022). How does the Adoption of Digital Technologies 652 Affect the Social Sustainability of Small-scale Agriculture in South-West Germany? 653 International Journal on Food System Dynamics. Scopus. 654 https://doi.org/10.18461/ijfsd.v13i3.C3 655 Wolfert, S., Ge, L., Verdouw, C., & Bogaardt, M.-J. (2017). Big Data in Smart Farming – A 656 review. Agricultural Systems, 153, 69–80. https://doi.org/10.1016/j.agsy.2017.01.023 657 Zhou, Y., Tiemuer, W., & Zhou, L. (2022). Bibliometric analysis of smart livestock from 1998-658 2022. Procedia Computer Science, 214, 1428–1435. 659 https://doi.org/10.1016/j.procs.2022.11.327. 660