Clasping technological advancement is a necessity of the modern era, specifically in Indian dairy. Researches are advancing towards ‘dairy 4.0’ and showcasing the optimisation of robotics, 3D printing, Artificial Intelligence, the Internet of Things, Big Data, blockchain in the milk industry (Abdo Hassoun et al., 2023). Although, irrespective of the utility, dairy stakeholders still rely on the conventional processes in production and management. A reason behind the resistant approach is fear of challenges and change, as innovation adheres to adaptation. The closer the dairy sector moves towards digitalisation, the harder it becomes to adapt, precisely, the unorganised sector, which accounts for nearly 62% of the established market. Where it must be noted that the unorganised sector is subsumed with ill-managed livestock, minimally skilled labours and unpaid aiders. This extends concern towards livestock management and sustainable production, along with the operations of the supply chain. Researchers have tried to address the mentioned issue by critically reviewing the present adaptation of digitalization in dairy with the help of precision livestock farming (PLF) technologies, that uses biosensors and data monitoring in livestock management (Neethirajan, 2023). The outcome of PLF technologies was positive, but the resistance towards its exploration turns out to be negative regarding the entire dairy industry. Even developed countries find it difficult to catch up with the pace of digital clock in the dairy sector, which makes ‘advance technology’ an unexplored realm in the dairy sector. To counter the same, the following study focuses on a bibliometrics approach, that includes in-depth analysis of present literature with the statistical tool, ‘Biblioshiny’. Sincere attempts have been made to highlight the contemporary functionality in dairy segment, concerning mechanisation and digitalization. By analysing the modernisation and technological adoption in dairy, present study magnifies its utility in encapsulating the Sustainable Development Goals. The various SDGs that could be achieved with the help of digital transformation are ‘Zero Hunger’, ‘Good Health and Well-being’, ‘Responsible Consumption and Production’ and ‘Sustainable Consumption’. However, the proportional relation of digital advancement and sustainable development is potentially unfathomable and the entire analysis endeavours the aforementioned.
 
Literature review
 
Agriculture-based industries are significant within the Indian business environment, with many skilled and semi-skilled workers engaged in dairy and agriculture. Integrating conventional skills with high-efficiency tools and machinery can yield substantial benefits. Literature suggests a positive response to this integration, particularly through technologies like blockchain, smart farming and big data analysis. Farmers are keen on big data for better decision-making, though there are challenges with data security, especially among semi-skilled workers (Newton et al., 2020). The Indian dairy sector has seen improved outcomes through the ISM-MICMAC approach, linking technology to enhance cattle production, herd management and efficient technology use (Kaushik and Rajwanshi, 2023).
       
The use of bio-sensing tools for real-time monitoring of livestock health is crucial, supported by government initiatives such as the national agricultural innovative project (NAIP), dairy entrepreneurship development scheme (DEDS) and Rashtriya gokul mission (RGM), which focus on technological advancement and social welfare in the dairy industry. Digitalization is essential not only for economic reasons but also for addressing broader social issues, as over 150 million households are involved in global dairy production.
       
Furthermore, digitalization in the food industry can help counter challenges like epidemics, war and economic instability. Researchers aim to transition the dairy sector to Industry 4.0, despite concerns about perishability and fraud (Malik et al., 2024). Technologies such as IoT and blockchain are vital for reducing fraud and managing supply chains effectively (Kumar et al., 2022; Himu et al., 2024). Overall, the push for digital transformation in Indian dairy is crucial for modernizing the sector and addressing ongoing challenge.
       
In India, there is a strong emphasis on strengthening the supply chain through milk cooperatives by incorporating smallholder farmers into state milk federations. In addition to focusing on supply chain coordination and utilising technology at milk unions and logistics, the Dairy Development Board has recently initiated basic e-consultancy and services for farmers through technology (Chaturvedi, et al., 2024).
       
Research indicates that encouraging the growth of organised dairy farms can help meet the anticipated future increase in demand for milk products both domestically and internationally.
       
In contrast to Asian countries, the Western world has well-established dairy farms that utilise technology in logistics and supply chain processes to deliver products from farms to consumers. The Indian dairy structure comprises milk cooperatives, unorganised local suppliers and self-consumption. This reliance limits the adoption of farm technologies, hinders expansion and restricts financial recognition for the organised sector’s contribution to GDP (Kaur and Toor, 2024).
       
The evolution of dairy farming is a testament to the remarkable progress from Industry 1.0 to Industry 4.0. Dairy 1.0 marked the beginning of this journey with fundamental mechanization, where reliance on manual labor was still prevalent. Mechanical milking machines and basic refrigeration systems set the stage for future growth, yet many traditional farming practices remained entrenched.
       
With the transition to Dairy 4.0, the landscape has transformed dramatically by integrating cutting-edge digital technologies like robotics, 3D printing, big data, IoT, AI and blockchain. These advancements have revolutionized the industry, driving significant sustainability, efficiency and productivity improvements. Automated milking systems enhance operational efficiency, reduce labor demands and minimize animal stress.
       
Moreover, 3D printing and big data have streamlined the production of machinery and enhanced farm management practices, while AI empowers farmers with predictive decision-making capabilities. Blockchain technology is a game changer for milk traceability, boosting customer trust and ensuring food safety. The healthcare approach in dairy farming has evolved, too, focusing on proactive disease management supported by IoT and AI (Sangode, 2025).
       
This research aims to explore the various literature available in this field. It also provides insights into future areas of study and trend analysis for researchers. To achieve these objectives, based on a review of existing literature on the theme of Artificial Intelligence, the following goals have been established:
1. To review the most cited authors and which are the most cited articles.
2. To identify countries that have contributed the most to this area of research.
3. To identify the trending topics over the period of years.
4. To examine the themes that are widely studied among academics.

The secondary based database was collected from the digital library of Harcourt Butler Technical University Kanpur (U.P.), in between the time period of July 2024 to October 2024.
       
Over the past two decades, Scopus has emerged as a leading database for abstracts and citations, providing quantitative analyses of academic literature. The field of bibliometrics analysis has been enhanced through various tools, including citation counts, keyword analysis, co-citation counts, co-occurrence measures and bibliographic coupling (Lwesya and Achanta, 2022). This paper aims to develop techniques for extracting data from both published and unpublished articles using VOSviewer for visualisation and Biblioshiny applications. Researchers prefer these applications because they offer a secure web interface for data importing, conversion, gathering and filtering from sources like Scopus. The primary focus is on identifying research papers related to the keyword phrase “Artificial Intelligence” within the multidisciplinary databases of Scopus.
       
The systematic literature review (SLR) involves three main steps: Data collection (which includes data loading and conversion), analysis and the development of descriptive and network matrices (co-citation, collaboration and co-occurrence). The final step is visualisation, which enables the creation of result maps. Relevant articles cited in the SLR contribute to a bibliometrics overview. Co-citation is defined as the occurrence of two documents being cited together by other documents (Small, 1973). The greater the number of co-citations that two documents receive, the stronger their co-citation relationship, indicating a higher likelihood of semantic connection. In this context, counting the frequency of keyword appearances allows for the analysis of research hotspots within disciplines (Huai and Chai, 2016). Co-authorship is a common practice in academia; thus, papers with multiple authors serve as indicators of collaborative activity within a field (Wagner and Leydesdorff, 2005). Table 1 shows the bibliometrix information of database.


 
Synthesis and analysis
 
The purpose of this paper is to conduct a descriptive and quantitative analysis of the role of Artificial Intelligence in dairy farming over the past twenty years, specifically from 2000 to 2020. Scopus, a citation and abstract platform established in 2004 and owned by Elsevier, has compiled an extensive collection of bibliographic databases, citations and references related to scientific publications across various fields, including science, technology, humanities and social sciences.Subsequently, each of the aforementioned categories are thoroughly analysed using the following elements (1) The document type, (2) Author impact analysis (3) Citation analysis (4) Trend analysis , (5) Word frequency over period of time (6) Word cloud (7) Co-occurence and (8) Thematic analysis.
       
Fig 1 shows author local impact by H-index and Fig 2 shows most cited authors in documents using vosviewer.




 
Author impact analysis
 
Refer to Fig 1, among the thirty authors in the database, the top ten authors who contributed the most to research on AI and modernisation in dairy farming have been identified. The complete publications of the three most productive authors during this period are from India, the United Kingdom and Italy. Researchers previously observed significant contributors who collaborated on multiple documents to identify the shared objectives of their partnerships. This approach is the most common method of investigating academic relationships in systematic research. The study was also conducted with the assistance of Biblioshiny. It was found that there are several strong co-author associations, with some pairs having collaborated on four publications together see in Fig 3 Cited Analysis for most cited countries.

​
 
Citation analysis
 
The term “citation,” also known as a bibliographic reference, refers to a formal acknowledgment of sources in research. Citation analysis is the study of how these references appear in bibliographies, lists or catalogs, particularly from sources outside the original context. This analysis is a commonly used bibliometric method that involves constructing citation graphs and networks, allowing researchers to explore the impact within their academic fields. Citations serve as a primary link between two documents: the one being cited and the one doing the citing. They represent a unique relationship between the cited and citing papers, establishing a connection between authors whose work can be measured through literature. In citation analysis, the commonalities among associations are based on subject areas, methodologies used and research domains. The main reasons for citing documents include acknowledging previous work and providing critical commentary on it. In this paper explains the most cited top ten countries in publishing articles on artificial intelligence in dairy farming. The top most is Australia with forty five citations in this domain (Fig 3).
 
Trend analysis
 
The trend analysis is illustrated in Fig 4, which shows the distribution of publication years from 2022 to 2024. The analysis highlights that topics related to “Artificial Intelligence” are expected to have the highest frequency in 2024. Other topics covered in the literature include milk production, machine learning, dairy farming and agriculture related to these domains. The shift towards incorporating Artificial Intelligence (AI) in various operational and supply chain activities within the dairy farming sector presents both new opportunities and challenges. Additionally, AI-enabled technology is proving to be an effective solution for avoiding disruptions in the supply chain and minimising risks. It generates positive outcomes, leading to the development of more solutions for logistics and supply chain issues. It is evident that digital dairy farming practices will continue to evolve and grow over time (Fig 4).

​
 
Thematic analysis
 
A thematic map is organised into four quadrants, divided along the x and y axes, which represent centrality and density, respectively. Co-occurrence network clusters are visually represented as bubbles on a graph. This innovative visualisation technique is based on Callon’s centrality and density rankings. Each bubble varies in size, reflecting the frequency of word occurrences within its cluster, which provides valuable insights into the prominence of specific themes in the research landscape. Each quadrant illustrates a different type of theme. The first quadrant focuses on niche themes related to animal husbandry and animal production, emphasising their relevance in terms of centrality. The second quadrant highlights main themes such as decision-making and future food supply, which are significant regarding development density. This quadrant includes prominent themes like artificial intelligence and milk production. The third quadrant identifies emerging and decoding themes, featuring subjects like dairies and the adoption of digital technology. Lastly, the fourth quadrant presents foundational themes associated with modernisation and human health. Both the first and third quadrants share common themes, such as dairying, dairy cows and milking machines, which have a notable degree of centrality. In contrast, artificial intelligence and milk production exhibit a low development degree yet possess a high degree of centrality (Fig 5). 

​
 
Tree map analysis
 
The TreeMap below (Fig 6) illustrates the combination of potential keywords related to Artificial Intelligence and modernisation in dairy farming. This milk production and agriculture theme is also prominent to address the issue of the article.

​
 
Word frequency
 
Word frequency is the intensity that author keywords refer to a selection of terms selected by authors to represent the relevant content of their articles. The frequency of the keywords is a valuable tool for researchers, search engines and indexers to discover relevant studies. The five most commonly used author keywords in this field are Artificial Intelligence, Agriculture, dairy farming and machine learning (as illustrated in Fig 7). Artificial Intelligence, an emerging concept, gained importance in the database in 2000 and has since risen to prominence as the most frequently used keyword in a database spanning from 1956 to 2024. Furthermore, while Digitalisation was recognised as an important keyword in the past, its significance has notably increased since 2015, Fig 7 shows word frequency over time (Source: Authors elucidation using biblioshiny and Fig 8 shows occurrence of most cited words.

​

​
 
Word cloud
 
The most frequently visualised keywords, presented in Fig 9, were generated using the Biblioshiny application interface in R-studio. This application extracts specific keywords based on the co-occurrence of author keywords, helping to identify the most significant and influential keywords based on their popularity. The top 20 keywords, which are strongly linked and high-impact, are more likely to be cited by future researchers compared to less-cited keywords. Important keywords such as “Artificial Intelligence,” “milk production,” and “Agriculture” should be explored alongside “modernisation,” “female animals” and “dairy farming” to uncover new insights and relationships. Countries like India, Australia and Italy connect to the “Dairy Farming” keyword. Bibliometric analysis is a visualisation tool for identifying current and future research trends (Pesta et al., 2018) Fig 9 shows the word cloud of most frequent keywords [Software: Biblioshiny (R-studio)].

​
 
Co-occurrence
 
Content co-occurrence analysis involves studying the content of publications, which can include topic areas such as titles and abstracts, as well as metadata like keywords. Literature reviews that utilise publication content allow researchers to distill various findings and contributions. This type of analysis enables the extraction and identification of theories, methods, contributions, samples, contexts, trending themes, highly cited themes, concepts and other valuable research data.
       
Using readily available software that employs algorithms to analyse data, such as VOSviewer and Bibliometrics, researchers can easily create visual representations of scholarship through co-occurrence maps. These maps demonstrate both inter- and intra-disciplinary research by revealing clusters or research streams related to the selected topic and dataset seen in Fig 10 Co-occurrence of sources in database using VOSViewer.

​
 
Bradford’s law
 
Bradford’s law is a vital concept that illustrates the distribution of articles on specific subjects across various periodicals. Since its introduction in 2006, the term “information scattering” has been recognised as a key phenomenon associated with information collections in our increasingly complex information landscape. The pattern of scientific articles dispersed in journals aligns remarkably with Bradford’s Law. When we categorise journals such that each category contains an equal number of articles on a given topic, the number of journals in each subsequent group typically follows a geometric progression.
       
To align perfectly with this law, certain criteria must be met: The bibliography should be comprehensive, cover a specific time frame and target a clearly defined subject. Despite this, the law often holds true even when these conditions are not fully satisfied. Brookes provides an engaging discussion on the implications of this law within library systems, emphasising its relevance. Currently, there is no simple model that clarifies the mechanics behind this law, high-lighting the importance of further exploration in this field.
       
Bradford’s Law describes how articles on a specific subject are distributed throughout various periodicals. Since 2006, the term “information scattering” has been used more generally to refer to a common phenomenon associated with information collections. The distribution of scientific articles in journals closely aligns with Bradford’s law. When journals are grouped such that each group contains the same number of articles on a particular subject, the number of journals in each succeeding group follows a geometric progression.
       
For strict conformity with this law, certain conditions must be met: the bibliography should be complete, cover a limited time frame and it can be easily shown that a frequency distribution J(p) of the number of journals with p articles follows the form:

J(p) ∝ p^(-r) (1)

 
Seen in Fig 11 which shows the diagrammatic representation of bradford’s law. 

​

From our extensive findings and bibliography, we can draw significant conclusions: Articles published in scientific journals are the most utilised format for sharing research results. In the past five years, over 60% of published papers have appeared in established databases. However, it’s important to note that many authors contribute only a single article, reflecting a lower than expected average productivity.
       
Our co-authorship analysis reveals a compelling trend: the majority of articles are written by two or three authors, leading to a co-authorship index of 2.1. Remarkably, nearly 90% of these authors are affiliated with prestigious universities in countries such as Australia, the USA, India and Italy. This indicates a strong collaborative effort in research that is pivotal for driving advancements in the dairy industry, particularly in the context of digital transformation.
       
Scientific journals remain the primary vehicle for disseminating research findings, with over 60% of papers published in databases in the last five years. However, it’s noteworthy that most authors have contributed only a single article, pointing to a low average productivity rate. Our co-authorship analysis underscores that most publications are indeed the result of collaboration between two or three authors, reinforcing the co-authorship index of 2.1. The institutions representing these authors predominantly include universities from Australia, the USA, India, Italy and similar countries.
       
These insights align with, which emphasizes that a small number of journals dominate the landscape of articles on specific topics. By harnessing the keywords derived from our research, we can effectively identify and connect with prior studies in our focus area of digitalization within the dairy industry, paving the way for future innovations and collaborations.

Our analysis is detailed in the literature review on the dairy sector focussing on   distinctive and comprehensive search terms on the subject. This process resulted in the retrieval of 31 documents from the Scopus database over the last five years, following our refinement procedures. It should be noted that using different keywords could have yielded a significantly larger number of results. Additionally, the parameters established to select the relevant papers present another limitation of this study. The research relies on a single database for data collection, which further constrains the findings.

The current research benefited immensely from the contributions of Sakshi Shukla and Dr. Chandra Kumar Tewari from HBTU Kanpur, who played a pivotal role in crafting the introduction and literature review sections. Additionally, under the expert guidance of Dr. Anand Kishore Chaturvedi from RTU Kota, Rachna Singh successfully authored the results and completed the remaining sections of the article.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.
 
Informed consent
 
No human and animals participated or involved in this study.

The authors declare that there are no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish or preparation of the manuscript.