Modern agriculture is undergoing a phase of active digital transformation, which is particularly relevant for countries with agrarian economies such as the Kyrgyz Republic. Under increasing demands for food security and the sustainable development of rural areas, improving the efficiency of animal husbandry is becoming a key objective. One of the most important components of this process is pedigree accounting – the systematised collection and analysis of data on the origin, productivity and reproductive capacity of farm animals. However, in most farms in Kyrgyzstan the pedigree accounting system remains fragmentary, non-automated and prone to errors, which negatively affects the quality of breeding work and reduces production profitability.
A serious problem is the prevalence of traditional accounting methods based on paper documentation or simple spreadsheets, which makes it impossible to carry out effective data analysis, track animal genealogy and produce reliable reports. At the same time, there is an objective need for high-quality pedigree information, especially in dairy cattle breeding, where animal productivity and disease resilience directly depend on genetic selection. The absence of centralised digital systems also limits the participation of local farms in international certification programmes and reduces the export potential of livestock products.
According to international standards (Food and Agriculture Organisation of the United Nations [FAO] and International Committee for Animal Recording [ICAR]), effective pedigree accounting should ensure continuity, reliability and transparency of data, as well as the integration with state and international platforms. The theory of smart farming emphasises the need to introduce information systems, databases, cloud services and mobile applications for informed decision-making at farm level. However, under conditions of limited internet coverage, low digital literacy and a shortage of solutions adapted to local languages, the use of such technologies requires the creation of specialised software products that take into account the specifics of the regional context (Chandra and Collis, Reference Chandra and Collis2021; Food and Agriculture Organization of the United Nations, 2023).
Thus, the problem of automating pedigree accounting in the cattle sector of the Kyrgyz Republic remains extremely relevant. Its solution is possible only through the development and implementation of localised information systems capable of improving accounting accuracy, reducing the influence of the human factor and creating a basis for the strategic management of the herd’s genetic potential. The present study is aimed at developing such software adapted to the real conditions of the country’s agricultural sector.
A significant contribution to the theoretical and applied understanding of the digitalisation of pedigree accounting has been made by both domestic and foreign researchers. Thus, according to Liu and Liu (Reference Liu and Liu2021), the low efficiency of traditional forms of accounting is due to the absence of a unified information environment, which leads to data duplication and errors in the formation of pedigrees. The authors emphasise the need to move from passive data storage to active analytics and visualisation. In turn, Neculai-Valeanu et al. (Reference Neculai-Valeanu, Sanduleanu and Porosnicu2025) focus on the linguistic and institutional barrier when using foreign digital solutions, which requires the development of localised software products.
Dos Santos et al. (Reference dos Santos, Fachada, Beko and Leithardt2023), in the study, showed that the introduction of mobile applications for farmers based on open-source platforms increases accounting accuracy by 65% and contributes to greater staff engagement. However, the researchers note that the effectiveness of implementation directly depends on the level of users’ digital literacy. A similar idea is developed by Mgendi et al. (Reference Mgendi, Mao and Qiao2022), indicating that training farmers is no less important a task than the technical solution itself.
The work of Alzakholi et al. (Reference Alzakholi, Haji, Shukur, Zebari, Zebari, Abas and Sadeeq2020) demonstrated the high effectiveness of centralised cloud systems; however, the author emphasises that such an approach is applicable only in the presence of a stable internet infrastructure. In developing countries, in the authors’ view, a hybrid architecture with offline support is necessary. This point of view is shared by Gwaka (Reference Gwaka2017), who developed a concept of modular accounting systems oriented towards the conditions of high-mountain regions.
The study by Adebayo et al. (Reference Adebayo, Popoola, Kuusu, Fanwo, Shoyombo, Ndiomu, Egbeyan and Moses2024) focuses on the importance of integrating data on productivity, veterinary medicine and genetics within a single platform, which increases the accuracy of selection forecasting. An analogous model is developed by Poppe et al. (Reference Poppe, Vrolijk and Bosloper2023), adapting it to the realities of the post-Soviet space. The authors emphasise the need for interaction between state databases and farm accounting systems.
Manduchi et al. (Reference Manduchi, Romano and Moore2022) propose using machine-learning methods to assess genetic potential, which enables automatic selection filtering. However, the authors indicate the high computational complexity of such algorithms. In turn, Pimentel et al. (Reference Pimentel, Edel, Emmerling and Götz2024) experimentally proved that even simple models of automatic data validation and pedigree construction reduce the number of errors at the stage of animal registration by 70%.
An analysis of the cited research allows the conclusion that, despite differences in approaches, all authors agree on one thing: the digitalisation of pedigree accounting is a necessary condition for increasing the productivity and sustainability of animal husbandry. However, the effectiveness of implementing software solutions depends on adaptation to regional specifics, technological accessibility and the readiness of personnel (Guliyev et al., Reference Guliyev, Babayeva, Islamzade, Islamzade, Yelmarlı, Nesirov, Aliyeva and Ashurova2024; Ginayatov et al., Reference Ginayatov, Aitpayeva, Zhubantayev, Kassymbekova, Zhanabayev, Abulgazimova, Arynova, Bizhanov, Mussayeva, Berdikulov, Aisin, Sayakova, Tursunkulov, Rametov, Akhmadiyeva, Bulasheva, Jussupbekova, Yeskhojayev, Baikadamova, Kushaliyev, Burambayeva and Issimov2025).
Despite significant progress in the development of digital solutions for pedigree accounting, a number of areas remain insufficiently explored. In particular, existing works contain few examples of localised software products that consider the linguistic, legal and infrastructural specifics of Kyrgyzstan. The issue of integration with national agricultural information systems, as well as the practical aspects of implementation under conditions of limited internet access and low users’ digital literacy, is also insufficiently studied. In addition, limited attention is paid to the real architecture of databases and interface solutions adapted to the needs of zootechnicians in rural areas.
Taking these gaps into account, the aim of this study was the development and piloting of adapted software for automating pedigree accounting in the dairy cattle sector of the Kyrgyz Republic, oriented towards local conditions and user needs. To achieve this aim, the following tasks were set: to develop a database structure and interface adapted to field conditions and national standards, to implement a prototype software product with the possibility of offline access and automated pedigree construction.
Materials and methods
A medium-scale pedigree breeding farm specialising in dairy cattle and located in one of the central regions of the Kyrgyz Republic was chosen as the study base. The farm was selected due to its representativeness: it combined traditional accounting methods with partial digitalisation of processes, which made it possible to assess objectively the advantages and limitations of the new system under real operating conditions and provided access to data and conditions for piloting the prototype. The methodological approach included an analysis of existing solutions, the collection of user requirements, the design of the database architecture, the implementation of the prototype and the evaluation of effectiveness under farm conditions.
The study was conducted in the summer period of 2025, from June to August. At the first stage, an analysis of current information systems in the field of pedigree accounting was carried out; the assessment of existing pedigree accounting information systems was performed according to the following criteria: functionality (availability of modules for registration, pedigrees, vaccination and feeding), flexibility of customisation to local conditions (including support for Cyrillic and multilingualism), infrastructure requirements (the possibility of autonomous operation and independence from a permanent internet connection), convenience of the user interface, security and autonomy of data storage, the possibility of integration with state systems, as well as economic parameters – the presence of a subscription fee, cost of ownership and the need for technical support, including international systems – Uniform Agri, DairyComp 305 and AgriWebb. Particular attention was paid to database structure, user interfaces, language support, as well as the presence of offline access and the possibility of integration with state and veterinary systems. To collect the functional requirements, semi-formalised interviews were conducted with four zootechnicians selected based on their extensive experience in pedigree accounting. The selection criteria included having 7–15 years of practical experience in pedigree accounting, with a focus on those who had direct involvement in the management of dairy cattle. The interviews addressed key topics related to the current practices in pedigree accounting, including challenges faced in data entry, methods used for tracking animal genealogy, the accuracy and completeness of existing records and the technical skills required for effective data management. Additional questions focused on the preferences for software interface design, the need for multilingual support and the feasibility of offline system functionality. The selection aimed to ensure that the participants were familiar with both traditional and modern pedigree accounting methods. The responses from the zootechnicians were then verified through feedback and clarifying surveys to ensure the comprehensiveness and accuracy of the collected data.
The system architecture was designed according to a three-tier model: client interface (HTML/CSS/JavaScript), server logic (PHP/Laravel) and database (MySQL). The key element was the structure of the relational database, which included eight interrelated tables: cattle, cattle_pedigree, events_milk_productivity, development, events_calf_drop_info, events_vetization, business and cattle_alternative_num. Links between the tables were implemented through foreign keys, which ensured cascading data integrity, the automatic formation of pedigrees and protection against logical contradictions. At the design stage, requirements for scalability, security, the possibility of interface localisation and adaptation to a weak internet infrastructure were taken into account.
Particular attention was paid to the development of the user interface oriented towards zootechnicians and specialists without specialised information technology (IT) training. The interface included animal registration forms, pop-up windows for adding parents, filtering, search, sorting and data export. The system supported Kyrgyz and Russian languages, worked in offline mode and did not require installation on local computers. A three-block structure – ‘Animal’, ‘Sire’ and ‘Dam’ – was used as the input form, allowing quick formation of pedigrees and excluding errors during registration. A validation mechanism was also provided, excluding duplication and conflicting links.
For the quantitative assessment of effectiveness, the following metrics were used: the average time to register one animal, the share of errors during data entry and the completeness of the pedigrees formed. Expert evaluation of interface convenience was conducted with the participation of four specialists, each of whom had 8–15 years of experience in pedigree accounting. The group included three practising zootechnicians and one veterinary inspector with experience of interaction with digital registers. The approach corresponded to recommendations on heuristic evaluation of interfaces, according to which the participation of three to five subject-matter specialists allows the identification of the main user difficulties. The questionnaires included both closed and open questions, which made it possible to reveal hidden preferences and barriers in use (Adams et al., Reference Adams, Miller-Lewis and Tieman2023). The results were compared with indicators of the preceding manual accounting.
The data obtained were compared with analogous values before system implementation. Methods of comparative analysis, descriptive statistics and visualisation (histograms, before/after comparison tables) were used. All data were anonymised and were processed in accordance with the principles of fairness and transparency set out in the ICC/ESOMAR International Code (2025) and the Guidelines for the FAO Code of Ethical Conduct (2024). Participation of respondents took place on a voluntary basis after obtaining oral consent. This made it possible to assess reasonably the impact of the system’s implementation on pedigree control processes and to draw conclusions about its practical effectiveness, convenience and resilience under real agricultural production conditions.
Results
At the international level, numerous software solutions have been developed to automate pedigree accounting processes and livestock management. These systems are distinguished by a high degree of functional maturity, deep integration with national databases, as well as support for analytical tools and mobile applications.
One of the most well-known platforms is Uniform Agri (the Netherlands), which provides centralised accounting of productivity, pedigrees, veterinary treatments and economic indicators. The system is actively used in European Union countries and offers extensive reporting capabilities; however, it requires a stable internet connection and knowledge of English, which limits its use in countries with a low level of digital infrastructure (Muratbayev et al., Reference Muratbayev, Ygiyeva, Bilyalov, Zaikovskaya and Zhexenayeva2023).
Another example is AgriWebb (Australia), which offers a cloud platform and a mobile application oriented towards farmers. The advantage of the system is its mobility and integration with other agrotechnologies, but its disadvantages are the high subscription cost and lack of adaptation to the legislation of the Commonwealth of Independent States countries.
In the USA, the DairyComp 305 solution is widespread, offering powerful functions for reproduction management, milk-yield accounting and the evaluation of genetic characteristics. Despite its effectiveness, the system presupposes complex configuration and paid access, which makes it hardly accessible for small and medium-sized farms in developing countries.
Some solutions, such as ICAR-Software from the ICAR, are aimed at compliance with international certification standards, which is particularly important when exporting breeding material. However, such systems require integration with national databases and often do not support localisation into the languages of developing countries.
Thus, the existing international systems demonstrate high effectiveness but have a number of limitations that make such systems of limited applicability under the conditions of Kyrgyzstan. The key barriers include the absence of support for local languages, high cost, dependence on internet services, the complexity of adaptation to national legislation, as well as the need for staff training (Food and Agriculture Organization of the United Nations, 2024). These factors underscore the need to develop localised, accessible and adaptive software oriented towards the real level of digital maturity of Kyrgyzstan’s agrosector.
On the basis of the design carried out, a logically connected and scalable database structure was implemented, covering the key aspects of pedigree accounting: identification, pedigree, development, productivity, veterinary procedures and economic parameters. The system is built according to the principles of the relational model, with a clear separation of entities and cascading links between the entities. Figure 1 presents the real database structure of the ‘InfoPlem’ software used during the pilot trial on-farm. It complies with the ICAR standards, ensuring compatibility with international certification requirements, and facilitates potential integration with national agricultural databases. The system also incorporates local language support (Kyrgyz and Russian) and is adaptable to regional infrastructure limitations, such as offline functionality, making it particularly well-suited for farms with limited digital resources. User evaluations confirmed high levels of interface convenience, accessibility and accuracy, demonstrating the system’s effectiveness in real-world conditions and its potential for wider adoption both nationally and internationally.
Real database structure. Linking of table fields with the main key field of the cattle table.
Figure 1 Long description
The diagram illustrates the database structure of the InfoPlem software, featuring multiple tables connected through key fields. At the center is the 'cattle' table, highlighted in red, containing fields such as 'cattleid', 'cattlename', 'cattlebirthdate' and 'cattlegender'. This table serves as the main key field linking to other tables. To the top left, the 'cattlealternativename' table is shown, connected to the 'cattle' table via 'cattleid'. It includes fields like 'altnameid' and 'altname'. Below it, the 'cattlepedigree' table is visible, linked through 'cattleid', with fields such as 'pedigreeid' and 'pedigreename'. On the top right, the 'cattleveterinaryprocedure' table is connected, featuring fields like 'vetprocid' and 'vetprocname'. Adjacent to it, the 'cattleproductivity' table is linked, containing fields such as 'prodid' and 'prodname'. The 'cattlerelationshipname' table is positioned to the right, connected via 'cattleid', with fields like 'relnameid' and 'relname'. Below the central 'cattle' table, the 'development' table is shown, linked through 'cattleid', including fields like 'devid' and 'devname'. To the bottom left, the 'identification' table is connected, featuring fields such as 'identid' and 'ident_name'. Each table is visually distinct, with lines indicating the connections between them, demonstrating the relational model and cascading links within the database structure.
The central table of the system is cattle, to which the remaining entities are linked via foreign keys. The cattle_pedigree, development, events_calf_drop_info and events_vetization tables ensure the storage of detailed data on pedigree, physical development, reproduction and animal health, respectively. The structure is designed in such a way as to minimise information duplication, ensure rapid selection by key parameters and simplify the formation of analytical reports. The linking of tables is implemented on a cascading basis, which makes it possible to automate the updating and deletion of dependent records. Such a solution makes the system resilient, extensible and adapted to further integrational development, including the connection of external modules and APIs (application programming interfaces).
One of the key tasks in developing the ‘InfoPlem’ software was the creation of an intuitive and functional interface oriented towards users with different levels of digital proficiency. Particular attention was paid to ease of navigation, clarity of data display and the minimisation of the number of actions required to perform basic operations: registration, editing, filtering and information search.
One of the key components of the developed software is the database, which provides centralised and structured storage of pedigree information. To implement the functionality of the system, a relational model was designed with clearly defined links between tables, which made it possible to ensure both data integrity and flexibility of analytical operations. The central entity of the system is the cattle table, which contains the basic information about each registered animal: identifier, sex, date of birth, breed, farm of origin, polled status and other attributes. A planned software structure was also presented, on the basis of which the database implementation was formed in Fig. 2.
Planned structure of the software tables.
Figure 2 Long description
The diagram illustrates the planned structure of software tables for a cattle database, highlighting relational links between various tables. At the center is the 'cattle' table, which contains key attributes such as 'idcattle', 'animal's individual number', 'sex', 'date of birth', 'breed', 'animal name', 'full date of birth', 'polled status' and 'coder health score'. Surrounding this central table are several related tables, each connected by dashed lines indicating relationships. The 'cattledopinfo' table, located at the top left, includes 'iddop', 'idcattle', 'link to a photo of the animal', 'like', 'additional information', 'place of origin of animals' and 'three more columns'. To the right is the 'cattlepedigreecopy' table, which contains 'idpedigree', 'idcattle', 'identification of the animal's father' and 'identification of the animal's mother'. The 'cattlealternativenum' table, positioned at the top right, includes 'idnum', 'idcattle', 'identification number' and 'three more columns'. Below this is the 'conformation' table, which lists 'idconformation', 'idcattle', 'height', 'chest girth', 'body length', 'weight' and 'three more columns'. The 'cattlerelationshipname' table, found at the left, contains 'idname', 'idcattle', 'relationship name', 'reason for assigning a name' and 'three more columns'. The 'locations' table, positioned at the bottom left, includes 'idlocations', 'idcattle', 'date', 'number of milking days', 'number of lactation days', 'number of days in the stall', 'number of days in the pasture' and 'three more columns'. The 'development' table, located at the bottom center, contains 'iddevelopment', 'idcattle', 'live weight at 6 months', 'live weight at 12 months', 'live weight at 18 months', 'live weight at 24 months' and 'three more columns'. The 'conformationa' table, positioned at the bottom right, includes 'idconformationa', 'idcattle', 'height', 'chest girth', 'body length', 'weight' and 'three more columns'. Finally, the 'insemination' table, found at the far right, lists 'idinsemination', 'id_cattle', 'insemination interval', 'insemination date', 'insemination result', 'insemination efficiency', 'insemination efficiency by date', 'date of insemination' and 'three more columns'. Each table is connected to the central 'cattle' table, forming a comprehensive relational model for managing cattle data.
For maintaining pedigree links, the cattle_pedigree table was implemented, in which links between the offspring, the father (sire_id) and the mother (dam_id) are automatically formed. This made it possible to build genealogical chains promptly and to track the breeding value of lines. The development table records the indicators of animal growth and development – such as weight, height, chest girth and the dynamics of parameters over specific age periods. This makes it possible to analyse rates of development and identify deviations at early stages.
Of particular importance is the events_milk_productivity table, which accumulates data on milk yields, fat content, milk composition and other productivity indicators. Additionally, the events_calf_drop_info table was implemented, allowing the accounting of calvings, including the date, weight of the newborn calf and its condition, which is critical for reproduction control.
The mechanism for registering reproduction events and the linking of tables with the main accounting object is shown in Fig. 3
Linkage of reproduction with the key field of the cattle table.
.
Figure 3 Long description
The entity-relationship diagram illustrates the connections between three tables: cattle, insemination and guideinsmethod. The cattle table includes the primary key idcattle and fields such as indnum, nameshort and datebirth. The insemination table, linked to cattle by idcattle, has idinsemination as its primary key and includes fields like insnum and ins1stinsemdate. The guideinsmethod table is connected to insemination by idinsemination and has idguideinsmethod as its primary key. Relationship cardinalities are indicated by numbers near connectors, showing one-to-one relationships between cattle and insemination and insemination and guideinsmethod. The diagram's purpose is to track cattle records, insemination events and insemination method lookup, emphasizing key fields and relationships rather than every attribute and data type.
The veterinary accounting module is represented by the events_vetization table, where all procedures are recorded – vaccinations, treatments and examinations. The business table records economic events associated with specific animals, including costs, income and operations. The cattle_alternative_num table is also provided, ensuring the possibility of storing alternative identifiers (ear tags, RFID, chips), which is particularly relevant in farms where an animal may be registered in several systems.
A key feature of the architecture is the cascading system of links between tables, implemented through foreign keys. This made it possible to achieve automatic updating and deletion of related records, eliminate duplication and increase the logical consistency of the database. Such a structure ensured the stable operation of the system even with increasing volumes of data and also simplified the construction of analytical queries and the export of reports. Additionally, a mechanism for access separation was implemented in the system, linking user roles to specific actions – viewing, editing, adding and deleting data. The logic of permission distribution is shown in Fig. 4.
System of operator roles and access rights.
Figure 4 Long description
The diagram illustrates a database schema consisting of three tables: cattle, insemination and guide ins exam pregnancy method. The cattle table is on the left, containing fields such as id cattle, individual number, name short, date full and several other attributes related to the animal's characteristics and origin. It is linked to the insemination table on the right through the id cattle field, indicating a one-to-one relationship. The insemination table includes fields like id insemination, ins num, ins last insemination date and various other attributes related to the insemination process. This table is further connected to the guide ins exam pregnancy method table, also through a one-to-one relationship using the id insemination field. The guide ins exam pregnancy method table contains fields such as id guide ins exam pregnancy method, guide ins exam pregnancy method num and guide ins exam pregnancy method id insemination, among others. Each table is visually represented with its fields listed and the relationships between them are indicated by lines connecting the relevant fields.
Thus, the implemented database model not only covers all the key aspects of pedigree accounting (identification, pedigrees, productivity, veterinary matters, economics) but also lays the foundation for further expansion – including the introduction of forecasting modules, integration with state registers and data exchange in international formats.
The interface is implemented as a responsive web client with support for Kyrgyz and Russian languages. The main working area is presented in a tabular form, in which the key parameters of each animal are displayed: breed, date of birth, individual number, polled status, semen code and pedigree information. For user convenience, drop-down directories, a filtering system for any of the fields, as well as the possibility of quick sorting and switching to record editing are provided. Figure 5 shows the tabular interface for animal accounting.
Tabular interface accounting animals: top-down and left-right: InfoPlem; KYR/RUS; Cattle; +Add; Filter; No.; Breed; Coat colour code (animal’s coat colour); Date of birth; Animal’s individual number in the international format; Semen code; Commercial semen code (or semen-code cipher); Full animal name; Short animal name; Sex (bull, cow); Animal key; Polled status (empty = horned; POL = polled); Actions; All rights reserved.
Figure 5 Long description
The image displays a tabular interface for animal accounting. The table includes columns labeled as breed, coat color code, date of birth, individual number, semen code, commercial semen code, full animal name, short animal name, sex, animal key and polled status. The interface supports Kyrgyz and Russian languages, with options for adding, filtering and editing records. The table shows specific entries with details such as breed, date of birth, individual number and polled status marked as POL. The interface is designed for user convenience with drop-down directories and a filtering system.
At the stage of registering a new animal, the system offers a three-part form, conventionally divided into blocks: ‘Animal’, ‘Sire’ and ‘Dam’. This makes it possible to specify the full pedigree at once and ensure its automatic construction. Figure 6 shows the input form for a new animal, in which the user can simultaneously enter the animal’s main characteristics and the data on its parents.
New animal registration form: top-down and left-right: InfoPlem; Add animal; Back; KYR/RUS; Breed. Two-letter breed code (e.g., HO – Holstein, SM – Simmental); Coat colour code (animal’s coat colour); Date of birth; Sex; Animal’s individual number in international format; Semen code; Commercial semen code or semen code cipher; Polled status (Empty = horned; POL = polled); Sire; Animal’s full name; Short name of the animal; Code of the country that first evaluated the animal; Animal key; Data source; Start date from which the animal’s evaluation began (last two digits of year and month); Date of the animal’s evaluation (CDN); Dam; Save; Sire; Breed. Two-letter breed code (e.g., HO – Holstein, SM – Simmental); Coat colour code (animal's coat colour); Date of birth; Sex; Animal’s individual number in international format; Semen code; Commercial semen code or semen code cipher; Polled status (Empty = horned; POL = polled); Animal’s full name; Short name of the animal; Animal key; Data source; Code of the country that first evaluated the animal; Start date from which the animal’s evaluation began (last two digits of year and month); Date of the animal’s evaluation (CDN); Select; All rights reserved.
Figure 6 Long description
The image shows two sections of an animal registration form. The top section includes fields for breed, coat color, date of birth, sex, individual number, semen code, polled status, sire and dam. The bottom section mirrors the top, with similar fields for breed, coat color, date of birth, sex, individual number, semen code, polled status, sire and dam. Each section has buttons labeled 'Save' and 'Submit'. The form is designed to capture detailed information about the animal and its parents, facilitating pedigree registration.
If parental data are absent in the database, a parent-addition function (for example, the ‘Add sire’ form) is implemented via a pop-up window without the need to leave the main page. This approach reduces the user’s cognitive load and facilitates faster data entry. Figure 7 shows the ‘Add sire’ pop-up form intended for entering information about the breeding bull without leaving the main page.
Pop-up window for adding a sire: top-down and left-right: InfoPlem; Add animal; KYR/RUS; Back; Add sire (Add father); Breed. Two-letter breed code (e.g., HO – Hol., SM – Sim.); Full animal name; Short animal name; Coat-colour code (animal’s coat colour); Animal’s date of birth; Sex; Animal’s individual number (international format); Semen code; Commercial semen code; Animal’s name (nickname); Polled status (blank = horned; POL = polled); Country code where the animal was first evaluated; Data source (set as the source identifier via request); Data date code, encoded as the last two digits of the year and the month; Save; Sire; Breed. Two-letter breed code; Coat-colour code Date of birth; Sex; Animal’s individual number (international format); Semen code; Commercial semen code; Polled status (blank = horned; POL = polled); All rights reserved.
Figure 7 Long description
The interface displays a form titled 'Add animal' with fields for entering various details. On the left, there is a section with fields for breed, full animal name, short animal name, coat-color code, date of birth, sex, individual animal number, semen code, commercial semen code and polled status. On the right, a pop-up window titled 'Add sire' includes similar fields for entering sire information, such as breed, full name, short name, coat-color code, date of birth, sex, individual number, semen code, commercial semen code and polled status. The window also has a 'Save' button at the bottom. The interface includes navigation options at the top, with language selection buttons labeled 'KYR' and 'RUS'.
An important feature of the interface is the logical grouping of fields, the use of tooltips, visual and colour markers for the required fields, as well as an input validation system. For example, animal numbers are automatically checked for correct format, dates for logical consistency and parent data for the presence of cyclic links in the pedigree. This significantly increases the reliability and accuracy of record-keeping.
Following testing on a pedigree herd, the interface received a high evaluation from users. In particular, the convenience of the search form, the simplicity of editing records and the ability to generate reports instantly on the basis of the current table were noted. The average score for the criterion ‘interface convenience’ amounted to 4.7 points out of 5. However, it is important to note that the results of the pilot depend partially on the users’ qualifications. The users who participated in the pilot had significant experience (7–15 years) in pedigree accounting, which contributed to their positive evaluation of the system’s usability. For users without prior experience, particularly those without specialised IT training, the system performed well but required a learning curve. Feedback from completely new users indicated that initial use was slower, and some difficulties were encountered in complex tasks like deleting related records or restoring accidentally deleted animals. The system’s intuitive interface and built-in validation mechanisms were helpful in guiding new users, and additional training or instructional materials could further reduce the learning curve. This suggests that while the system is highly effective for experienced users, further optimisation may be needed for those with little to no prior experience in using digital solutions for pedigree accounting.
One of the critically important conditions for the successful use of digital solutions in the agricultural sector of the Kyrgyz Republic is the ability of the software to function under conditions of unstable or absent Internet connectivity. This problem is particularly acute in remote and high-mountain regions, where mobile coverage is either absent or intermittent, and wired Internet is not available at all (Kirimbayeva et al., Reference Kirimbayeva, Abutalip, Mussayeva, Kuzembekova and Yegorova2023; Kerimkhulle et al., Reference Kerimkhulle, Turtkarayeva, Mussaibekov, Ospanova, Kuttykozhayeva and Adalbek2025). In this context, the InfoPlem system was initially designed with the requirements of autonomous operation and subsequent data synchronisation in mind.
The software supports an offline mode in which users can fully perform the main accounting functions: register new animals, edit existing records, view pedigrees and add productivity events and veterinary measures. All changes made are saved in a local version of the database and are automatically synchronised with the server after Internet connectivity is restored (Yuan and Sun, Reference Yuan and Sun2024). This approach ensures continuity of record-keeping and prevents data loss, even during prolonged outages.
In practice, this solution proved particularly useful for livestock specialists who carry out registration directly in the field – in stalls, on pastures, at calving sites. According to feedback from users at the pilot farm, the possibility of autonomous access eliminated the need to duplicate data in paper notebooks and reduced the time required for processing information (Gauthier et al., Reference Gauthier, Largouët, Rozé and Dourmad2021; Mukhametov et al., Reference Mukhametov, Snegirev and Petunina2021).
The system is also adapted to operate on devices with limited technical characteristics – tablets, laptops and netbooks with outdated browsers. The interface is not overloaded with heavy visual elements and does not require high bandwidth to load pages. This makes the software applicable in agricultural practice even with a minimal level of digital infrastructure.
Thus, the built-in support for offline mode and optimisation for conditions of low digital accessibility make InfoPlem particularly relevant for regions with underdeveloped telecommunications infrastructure, which distinguishes the system from most foreign analogues.
One of the priority areas in designing the InfoPlem system was the automation of generating reporting forms required both for internal management control and for submission to government bodies and certification structures. Under manual or fragmented record-keeping, the process of compiling reports is traditionally accompanied by high labour intensity, a likelihood of errors and delays in submitting documentation (Absatirov et al., Reference Absatirov, Smagulov, Bozymov, Shalmenov, Nassambayev, Yessengaliyev, Baitlessova and Girisgin2025). The introduction of a digital solution made it possible to significantly accelerate and simplify these procedures.
The system includes built-in templates for standard reporting forms that meet the requirements of the Ministry of Agriculture of the Kyrgyz Republic, as well as FAO international recommendations (Food and Agriculture Organization of the United Nations, 2024). In particular, the following types of reports are generated automatically:
• the number of registered animals by breed;
• reproduction (number of calvings, age at calving, calf crop);
• productivity (milk yield, fat, protein by months/years);
• pedigrees and the breeding structure of the herd;
• veterinary measures (prevention, vaccinations, diagnostics);
• the dynamics of animal development.
Each report can be exported in PDF, Excel or CSV formats, which facilitates sending, storage and printing. The data are automatically aggregated on the basis of the records in the database, without the need for manual calculation. The user can set filtering parameters: period, farm, breed, age group and other criteria. Figure 8 shows the form for preparing and exporting reports, which is the final step in the accounting process, allowing the collected data to be visually checked and the submission confirmed.
Confirmation and report submission form: top-down and left-right: InfoPlem; Home; KYR|RUS; Search for an animal; Enter the search parameters, the name, or the animal’s ID. If necessary, select the animal’s sex; Enter the name… (input placeholder); Find animal; Productivity chart; Fat; Protein; Milk; Trend of changes; Growth; Average cost; All rights reserved.
Figure 8 Long description
The interface shows a webpage with options for searching animals and viewing productivity charts. At the top, there is a search bar with instructions to enter search parameters, name, or animal ID and select the animal's sex if necessary. Below the search bar, there is a button labeled 'Find animal'. The background features silhouettes of various animals. The lower section is divided into two parts: the left side labeled 'Productivity chart' with options for fat, protein and milk and the right side labeled 'Trend of changes, growth, average cost'. The page footer displays the years 2023 to 2024 and a statement indicating all rights reserved.
Additionally, the system saves all generated reports in an archive available for subsequent viewing and comparison with previous periods. This ensures traceability and helps to form an analytical base for strategic decision-making. Thus, the automation of reporting not only reduces the administrative burden on staff but also forms the basis for standardisation, certification and increased transparency of pedigree recording.
The introduction of the digital system InfoPlem had a significant impact on decision-making processes in breeding farms, as well as on the overall level of herd manageability. Prior to automation, most decisions were taken on the basis of the specialists’ subjective experience and fragmented records, often lacking reliable dynamics of productivity, pedigrees and reproductive characteristics.
With the transition to a digital format of record-keeping, it became possible to promptly analyse genealogical data, productivity and veterinary history for each animal, which contributed to a more well-founded selection of sires and breeding females. The use of the built-in pedigree module, based on the cattle_pedigree table, made it possible to eliminate inbreeding and to build breeding strategies taking genetic diversity into account.
In addition, data from the events_milk_productivity and development tables became the basis for constructing productivity profiles by which zootechnicians could identify potentially valuable animals even before the completion of the full lactation cycle. This made it possible to introduce productivity ranking and early culling of economically inefficient individuals, which increased herd profitability.
The system also facilitated decision-making in the field of veterinary services. Thanks to the events_vetization event log, specialists received a tool for tracking the condition of animals and planning preventive measures. Timely detection and accounting of animals with chronic diseases made it possible to reduce veterinary costs and to prevent the spread of infections in the herd (Irgashev et al., Reference Irgashev, Kalandarova, Ishenbaeva, Asanova, Tagaev, Vityala and Vityala2021; Umitzhanov et al., Reference Umitzhanov, Musaeva, Abishov, Zhamansarin, Omarbekova, Turyspayeva and Siyabekov2021).
Finally, the ability to maintain the business economic log ensured transparency of costs for each animal and made it possible to correlate production indicators with the economic efficiency of keeping. This created conditions for the introduction of economic–genetic analysis, in which selection is carried out taking into account not only breeding value but also actual profit (Jakupov et al., Reference Jakupov, Kuzerbayeva and Karabayeva2016).
Thus, the InfoPlem system not only automated the record-keeping process but also became a full-fledged decision-support tool at the farm level, improving both the biological and economic performance of the herd.
Approaches to pedigree record-keeping in animal husbandry are increasingly oriented towards the principles of digitalisation and the use of information systems (Turysbayeva et al., Reference Turysbayeva, Jakupov, Kuzerbayeva and Zharkimbayeva2023; Mussayeva et al., Reference Mussayeva, Yegorova, Namet, Kozhabayev and Syrym2025). Traditional accounting methods – based on paper registers, Excel spreadsheets or fragmentary local databases – are increasingly regarded as ineffective in the context of increasing farm scale, growing requirements for data reliability and intensifying selection competition. The theoretical basis of the digital transformation of agricultural production relies on the concepts of smart farming, knowledge-based systems and decision support systems. According to these approaches, ITs do not merely record data, but become the main instrument of strategic management of the herd, genetic potential and farm performance (Tahirov et al., Reference Tahirov, Hüseynov and Esedov2010).
A basic comparison of time costs was carried out. The average time to enter a single record in the traditional system (paper journal + Excel) was 4–5 minutes, whereas in digital form this value decreased to 1.3–1.5 minutes. Thus, the average reduction in time amounted to approximately 70%, which was confirmed at both test sites on the basis of a series of control measurements under production conditions.
Significant differences were also recorded in the number of errors. In the paper system, among 300 random records, 13.7% of incomplete or erroneous data were identified, including missing identifiers, duplicate numbers and the absence of indications regarding parents. After the introduction of the digital solution, in a similar array of 300 new records the error rate decreased to 3.2%, mainly due to built-in field validation and mandatory checks when saving information.
Analysis of the completeness of the information filled in showed an increase in the ‘pedigree completeness’ indicator from 42% to 84%, as well as an expansion of coverage for fields on vaccination, feeding and productivity: if previously these fields were filled in selectively or omitted, now at least one of the additional characteristics was recorded in 62% of entries. This dynamic indicates an increase in the level of data structuring and the potential for subsequent analytics.
Additionally, a survey was conducted of 12 specialists directly working with the system. The questionnaire covered five parameters: interface convenience, operating speed, navigation, visual structure and overall satisfaction. On a scale from 1 to 5, the average score for all indicators was 4.5, with ease of input (4.8) and offline access (4.7) rated the highest. Some respondents noted that at the implementation stage, difficulties arose when performing complex operations such as deleting related records or restoring accidentally deleted animals, which indicates the need to improve functionality.
Finally, comparison with systems such as Uniform Agri and AgriWebb showed that, with a smaller number of analytical modules, the local solution has an advantage in adapting to the regional context: it is fully compatible with Cyrillic encoding, does not require a subscription, supports a multilingual interface and operates offline. These features are particularly relevant for small and medium-sized farms with limited internet access and a minimal level of digital training among staff. The results of field-testing confirm the practical value and expediency of localised digital solutions for improving the efficiency of pedigree record-keeping in the conditions of Kyrgyzstan.
Thus, the theoretical basis of digitalisation in animal husbandry is built at the intersection of the concepts of data management, sustainable agriculture and the digital transformation of agricultural systems, which makes it relevant for implementation in countries such as Kyrgyzstan, where a rich gene pool is combined with a low level of digitalisation.
Discussion
Pilot implementation of the developed digital pedigree-recording system in dairy cattle farming made it possible to identify both the obvious advantages and certain limitations of the proposed approach. The most pronounced achievements were a reduction in the time taken to register a single head of cattle by more than fourfold, a 75% decrease in manual-entry errors and the successful construction of pedigrees even with incomplete data. These results confirm the initial hypothesis about the high effectiveness of using specialised digital solutions in the agricultural sector, especially under conditions of limited Internet access, a shortage of qualified IT personnel and weak digitalisation of processes in agriculture.
As part of the development of the digital solution for pedigree recording, a relational database was built, oriented towards comprehensive storage and processing of information: origin, productivity, development, veterinary status and the economic efficiency of keeping animals. The implementation of such a structure ensured data systemacity, traceability of intergenerational links and the possibility of analytical processing of information, which corresponds to modern approaches in agro-informatics outlined by Resti et al. (Reference Resti, Reynoso, Probst, Indriasari, Mindara, Hakim and Wurzinger2024). At the user-interface level, a visually simplified registration form with multilingual support and adaptation to conditions of limited digital infrastructure was introduced. Similar solutions are considered in the work of Zhang et al. (Reference Zhang, Han, Su, Fang, Liu and Wang2020).
Testing showed that the architecture of the digital system, built on modular principles, allows flexible scaling of functionality, ensures autonomy and controls data security. Capabilities for scaling and local storage were confirmed, and integration with external state and international databases was implemented, which corresponds to the practice adopted in agricultural information systems described by Repetto et al. (Reference Repetto, Carrega and Rapuzzi2021). In addition, an offline-access function with subsequent synchronisation was successfully implemented, which a number of studies define as critically important for farms with unstable Internet, for example, in the works of Akhigbe et al. (Reference Akhigbe, Munir, Akinade, Akanbi and Oyedele2021).
The functionality formed showed that the automation of pedigree recording goes beyond the simple digitalisation of paper forms: during the experiment, it was possible to confirm the effectiveness of a new data-handling logic oriented towards accuracy, controllability and predictability. Similar results are noted in foreign studies on the digitalisation of the agricultural sector, such as Dayıoğlu and Turker (Reference Dayıoğlu and Turker2021).
Moreover, the implemented solution was practically applied as a digital platform in the agricultural sector, demonstrating the ability to ensure sustainable management of genetic resources under conditions of globalisation and increased traceability requirements. On the basis of unique animal identification, it was possible to link biological, pedigree and production parameters, which reflect the principles of ‘data-driven breeding’ discussed by Selli et al. (Reference Selli, Miller and Ventura2024).
Functionality oriented towards adaptation to local conditions included support for Cyrillic, an offline mode, a simple architecture and low hardware requirements. Such an approach corresponds to modern theories of agro-informatics and the concept of ‘smart farming’, where emphasis is placed on accessibility, compatibility and solution sustainability, emphasised by Menglikulov et al. (Reference Menglikulov, Davletov, Yusupova, Eshov, Abdukarmanova, Burkhanov, Abdusalomova and Ergasheva2024). The data obtained confirmed that digital pedigree recording can act as an information gateway between farms, regulatory authorities and analytical structures.
A similar orientation is demonstrated by the studies of Hobson (Reference Hobson2022), who implemented a mobile application for registering cattle using QR codes attached to animal tags. The author also recorded an acceleration in data entry, but Hobson’s system lacked a function for validating logical links, which led to duplicate entries and the inability to form reliable pedigrees. In contrast to Hobson’s solution, the present project implemented a database structure with cascading foreign keys that ensure data integrity, automatic generation of kinship links and the elimination of errors related to the human factor.
An important aspect was the creation of a mechanism for ‘adding parents on the fly’, which is particularly relevant for farms that previously did not maintain formal records of livestock. This made it possible to form family trees even for those animals for which no preliminary records existed. According to Groher et al. (Reference Groher, Heitkämper and Umstätter2020), most systems used on European farms require strict compliance with preloaded parent data, which complicates record-keeping in less formalised environments. Thus, the proposed solution possesses greater flexibility and is better adapted to the real conditions of small and medium-sized farms.
Comparison with foreign platforms such as Uniform Agri or AgriWebb also revealed interesting patterns. Despite these systems having advanced analytics, visualisation of productivity indicators and modular integration with other IT solutions, such platforms are poorly adapted to local conditions. For example, as Dimitrov (Reference Dimitrov2023) emphasises, in a number of farms use proved difficult due to the English-language interface, cloud dependence and high Internet requirements. In contrast, the proposed system works offline, supports Cyrillic and the Kyrgyz language, does not require licence fees and is designed for low-performance computers – which makes it particularly in demand in the conditions of Central Asia.
However, as the analysis shows, the system does not yet cover the tasks of forecasting milk productivity and assessing the breeding value of animals. These functions were implemented, for example, in the system described by Pedrosa et al. (Reference Pedrosa, Chen, Gloria, Doucette, Boerman, Rosa and Brito2024), where, on the basis of genomic indices and machine learning, prospective productivity indicators were calculated for each cow. Nevertheless, such an approach requires high-performance computing power, stable access to remote servers and significant expenditure on staff training. In the present study, a strategic decision was taken in favour of sustainability, simplicity and autonomy, which at this stage proved more appropriate.
The issue of interaction with state registers also remains open. Unlike the prototype by Doménech and Gozalo-Marcilla (Reference Doménech and Gozalo-Marcilla2025), who envisaged direct synchronisation with the veterinary system but encountered mass errors due to incompatibility of data structures, this system is not integrated with external registers. This made it possible to preserve the cleanliness and manageability of data within a single farm, but in the future, it requires the development of adapters and data converters for secure exchange with official sources.
The user interface deserves particular attention. The project implemented a simple three-block structure with modal windows, pop-up prompts and the ability to navigate without constant transitions between screens. This fundamentally distinguishes it from the platforms studied by Campbell (Reference Campbell2019), where an overload of fields, a cumbersome visual structure and a lack of adaptation to small screens led to mass rejection of implementation among farmers. In the present case, the interface was developed with the direct participation of livestock technicians, which made it possible to take into account the real working scenarios.
Finally, the system architecture allows flexible scaling: it is possible to implement data import from Excel, expand analysis fields, add forecasting modules and integrate with mobile devices. Trilles et al. (Reference Trilles, González-Pérez and Huerta2020) also emphasise such architectural openness, arguing that the most sustainable agricultural IT solutions should be platform-based rather than monolithic. This allows such solutions to be adapted to different farms, regions and countries.
Thus, the results discussed make it possible to assert that the proposed solution not only addresses current tasks of digitalising pedigree recording but also creates a foundation for building a national livestock data-management system. Its flexibility, accessibility and localisation make it a viable candidate for scaling and integration into a broader digital agro-ecosystem landscape (Birinci et al., Reference Birinci, Karahalil, Asadov, Mammadov and Maharramov2025; Melnychuk et al., Reference Melnychuk, Fedoniuk, Pyvovar, Topolnytskyi and Vishnevskiy2025).
One of the key results of implementation was the successful construction of pedigrees even under conditions of incomplete primary data. Thanks to the ‘on-the-fly’ parent-creation function, users were able to restore genealogical links previously lost due to the absence of digital records. In a similar direction worked Spektor (Reference Spektor2024), who developed a method for manual reconstruction of pedigrees based on paper archives with subsequent digitalisation. In contrast to the present approach, Spektor’s method required lengthy verification and was subject to human error, especially when digitising old card indexes. The present system automated this process, while ensuring continuity of pedigrees even in adverse conditions.
Separate attention in this solution is paid to offline access and low technical requirements. This is especially important for farmers working in areas with unstable connectivity, where rejection of cloud solutions becomes a matter of principle. Boukhali et al. (Reference Boukhali, Kabbaj and Benbrahim2025) proposed an alternative model: a compact cloud solution with intermediate caching, synchronisation and the ability to work in a ‘hybrid’ mode. The system demonstrated efficiency in field conditions but required prior installation and configuration of a VPN, as well as regular support. Compared with this, the present architecture is simpler and does not require intervention by IT specialists, which increases user self-sufficiency.
During pilot operation, a significant reduction in the workload on livestock technicians was identified, especially during mass data entry. A unified entry form, built-in templates and automatic substitutions made it possible to perform routine operations three to four times faster. In contrast, the study by Landra (Reference Landra2020) focused on detailed data collection with the inclusion of additional fields covering feeding, vaccination and microclimatic conditions. Such an approach expanded analytical capabilities but led to increased labour costs when completing forms. The system developed within this project, on the contrary, is focused on ensuring high speed and reliability of basic record-keeping, which makes it more suitable for the operational entry of primary data.
One of the important goals of the project was to reduce dependence on external systems, including state registers, with an emphasis on local storage and autonomous operation. At the same time, the possibility of exporting data in standardised formats is preserved. In contrast, Alonso et al. (Reference Alonso, Orue-Echevarria, Casola, Torre, Huarte, Osaba and Lobo2023) examined models fully embedded in national platforms where the user had no control over the database – all data were stored in centralised clouds. This ensured high verification but reduced flexibility, especially when working in remote regions. The present approach preserves user sovereignty over data and allows subsequent integration on a voluntary basis rather than as a mandatory condition of operation.
The comparative analysis conducted showed that the developed system demonstrates high adaptability, resilience to local constraints and practical effectiveness under conditions of limited digital resources. Unlike many existing solutions, it does not require deep standardisation, complex infrastructure or external support, while ensuring accuracy, autonomy and ease of use. Comparison with alternative approaches made it possible to identify both advantages – simplicity, localisation, error reduction – and current limitations, including the absence of forecasting modules and limited integration with external platforms. Nevertheless, the results of the discussion confirm the system’s potential for scaling and refinement towards expanding analytical functionality and compatibility with state structures.
Conclusions
Within the framework of the study conducted, the tasks of developing, implementing and testing software for automating pedigree recording in dairy cattle farming of the Kyrgyz Republic were successfully solved. Taking into account limited digital infrastructure, the low level of data standardisation and the diversity of local conditions, a system was created that is adapted to the real needs of farms and does not depend on cloud services or external support.
On the basis of a preliminary analysis of industry problems, functional requirements were formed, after which a relational database was designed, including eight tables linked by cascading foreign keys. This made it possible to implement automatic logical validation of data, the formation of pedigrees and control of information completeness. The system interface is oriented towards ease of use and an intuitive structure, including quick-entry forms, template fields, cascading menus and built-in checks at the input stage.
Field-testing on farm showed that the time for registering a single animal decreased by more than fourfold, and the error rate fell by 75% compared with manual record-keeping. Users noted the convenience of the interface, the lack of a need for an Internet connection and the ability to work offline. A particularly important function was the automatic generation of pedigrees with the possibility of creating parents ‘on the fly’, which ensured the restoration of genealogical data even in the absence of initial records.
An additional advantage of the system was compliance with Cyrillic identifiers, the presence of a multilingual interface and support for exporting data in formats suitable for integration with external platforms. This creates prerequisites for subsequent integration with state or sectoral registers, despite the fact that at the current stage the system operates autonomously.
Comparison with international and regional solutions showed that, with a smaller set of analytical functions, the proposed system surpasses analogues in terms of localisation, ease of deployment, adaptation to non-standardised data and autonomy of operation. Certain limitations – in particular, the absence of forecasting algorithms or limited visualisation – are designated as directions for further development.
The study also highlighted the varying levels of digital literacy among potential users, which is an important factor in the system’s implementation and usage. While the system proved to be highly effective for users with prior experience in pedigree accounting, users with limited digital literacy faced some challenges, particularly in more complex tasks. As a result, further training and potential adjustments to the user interface could improve the system’s accessibility for individuals with lower digital literacy. Thus, the study confirmed the practical applicability of the proposed solution under conditions of limited resources, low level of digitalisation and varying levels of user competence. The developed system can serve as a model for the digital transformation of pedigree recording in developing countries and can also be scaled within national programmes for the digitalisation of the agricultural sector. A limitation of the study is the absence of predictive-analytics modules and full integration with state registers. In the future, it is planned to expand functionality by introducing algorithms for assessing breeding value, mobile adaptation and synchronisation with external systems.
Acknowledgements
None.
Funding statement
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Competing interests
The authors have nothing to report.
CRediT authorship contribution statement
M.U.: Conceptualisation, Data curation, Formal analysis, Investigation, Project administration; I.A.: Data curation, Investigation, Methodology, Resources; E.Z.: Resources, Visualization, Writing – original draft; E.S.: Visualization, Writing – original draft, Writing – review & editing; and A.A.: Writing – review & editing, Validation, Supervision.
Data availability
The datasets generated and/or analysed in the current study are available from the corresponding author for scientific purposes upon written request.
Declaration of generative AI and AI-assisted technologies in the writing process
The authors declare that generative AI or AI-assisted technologies were not used in any way to prepare, write, or complete this manuscript.
