Prediction of Lactation Milk Yield in Simmental Cattle Milked with Robotic Milking System Using CHAID and CART Algorithms

Öz

Animal husbandry has long been a key component of agriculture, fulfilling essential nutritional needs. Technological advancements have gradually replaced human labor with machines, particularly in dairy farming, where the milking process is vital for income generation. Robotic milking systems have emerged as significant innovations, allowing for efficient, hygienic, and automated milking while reducing dependence on labor.This study aims to predict lactation milk yield (LMY) in Simmental cows during their first lactation period in robotic milking farms by using various factors, including Days in Milk (DIM), Status (S), Number of Inseminations (IN), Milk Flow Rate (MFR), Robot Rejection Rate (RRR), Rumination Time (RT), Time Spent in the Robot (TSR), Feed Amount in the Robot (FAR), Feed Consumption Rate in the Robot (FCRR), and Milking Frequency (MF). The analysis incorporates Classification and Regression Trees (C&RT) and Chi-squared Automatic Interaction Detector (CHAID) algorithms, identifying DIM as the primary predictor.The CHAID analysis revealed that newly calved cows (DIM < 30) had an average LMY of 5,692 L, while those receiving over 5.09 kg of feed achieved an average of 8,426 L. For cows in the 30 to 81 days of lactation, higher feed allocation correlated with increased milk yield. The CART algorithm confirmed these findings, establishing DIM as the most influential factor. Overall, robotic milking systems facilitate individualized management of dairy cows, optimizing factors such as feed allocation and milking frequency. By leveraging advanced algorithms to analyze these variables, this study highlights the potential for improving milk yield and animal welfare in modern dairy farming practice.

Anahtar Kelimeler

• Robotik sağım
• Süt verimi
• Süt sığırı
• Veri madencilği
• CART
• CHAID

Robotik Sağım Sistemiyle Sağılan Simental Sığırlarının Laktasyon Süt Veriminin CHAID ve CART Algoritmaları ile Tahmini

Öz

Hayvancılık, uzun süredir tarımın temel bir bileşeni olup, temel beslenme ihtiyaçlarını karşılamaktadır. Teknolojik gelişmeler, özellikle sütçülükte, insan emeğini makinelerle kademeli olarak değiştirmiştir; burada sağım süreci gelir elde etmede hayati bir öneme sahiptir. Robotik sağım sistemleri, verimli, hijyenik ve otomatik sağımı mümkün kılarak, iş gücüne olan bağımlılığı azaltan önemli yenilikler olarak ortaya çıkmıştır. Bu çalışma, robotik sağım çiftliklerinde Simental ineklerinin birinci laktasyon döneminde laktasyon süt verimini (LSV), Days in Milk (DIM), Statü (S), Dölleme Sayısı (IN), Süt Akış Hızı (MFR), Robot Reddi Oranı (RRR), Ruminasyon Süresi (RT), Robotta Geçirilen Süre (TSR), Robotta Verilen Yem Miktarı (FAR), Robotta Yem Tüketim Oranı (FCRR) ve Sağım Sıklığı (MF) gibi çeşitli faktörleri kullanarak tahmin etmeyi amaçlamaktadır. Analiz, Sınıflandırma ve Regresyon Ağaçları (C&RT) ve Ki-Kare Otomatik Etkileşim Algılama (CHAID) algoritmalarını içererek, DIM'i birincil tahmin edici olarak belirlemiştir. CHAID analizi, yeni doğum yapmış ineklerin (DIM < 30) ortalama LSV'sinin 5.692 L olduğunu, 5.09 kg'dan fazla yem verilenlerin ise ortalama 8.426 L'lik süt verimi elde ettiğini ortaya koymuştur. 30 ile 81 gün arasındaki laktasyon dönemindeki ineklerde, daha yüksek yem tahsisi ile artan süt verimi arasında bir ilişki bulunmuştur. CART algoritması, bu bulguları doğrulamış ve DIM'i en etkili faktör olarak belirlemiştir. Genel olarak, robotik sağım sistemleri, süt ineklerinin bireysel yönetimini kolaylaştırarak, yem gibi faktörlerin optimizasyonunu sağlar. Bu çalışmada, bu değişkenleri analiz etmek için gelişmiş algoritmalar kullanılarak, modern sütçülük pratiğinde süt verimi ve hayvan refahını iyileştirme potansiyeli vurgulanmaktadır.

Anahtar Kelimeler

• Robotic milking; milk yield; cattle
• data minig
• CART
• CHAID

Kaynakça

1. Altay Y (2022a). Prediction of the live weight at breeding age from morphological measurements taken at weaning in indigenous Honamli kids using data mining algorithms. Tropical Animal Health and Production 54(3), 172.
2. Altay Y (2022b). Phenotypic characterization of hair and Honamli goats by using classification trees algorithms and multivariate adaptive regression splines (MARS). Kafkas Üniversitesi Veteriner Fakültesi Dergisi 28(3), 401-410. https://doi.org/10.9775/kvfd.2022.27163
3. Altay Y and Albayrak Delialioğlu R (2022). Diagnosing Lameness with the Random Forest Classification Algorithm Using Thermal Cameras and Digital Colour Parameters. Mediterranean Agricultural Sciences 35, 47-54.
4. Altay Y, Boztepe S, Eyduran E, Keskin I, Tariq MM, Bukhari FA, Ali I (2021). Description of factors affecting wool fineness in Karacabey Merino sheep using CHAID and MARS algorithms. Pakistan Journal of Zoology 53(2), 691. https://doi.org/10.17582/journal.pjz/20190329150359
5. Amos HE, Kiser T, Loewenstein M. (1985). Influence of milking frequency on productive and reproductive efficiencies of dairy cows. Journal of Dairy Science 68(3), 732-739.
6. Bach A, Cabrera V (2017). Robotic milking: Feeding strategies and economic returns. Journal of Dairy Science 100(9), 7720-7728. https://doi.org/10.3168/jds.2016-11694
7. Breiman L, Friedman JH, Olshen RA, Stone CJ (1984). Classification and regression trees. Chapman and Hall, Wadsworth Inc., New York, NY, USA.
8. Britt J, Cushman R, Dechow C, Dobson H, Humblot P, Hutjens M, Jones G, Ruegg P, Sheldon I, Stevenson J (2018). Invited review: Learning from the future—A vision for dairy farms and cows in 2067. Journal of Dairy Science 101(5), 3722-3741. https://doi.org/10.3168/jds.2017-14025

9. Coşkun G, Şahin Ö, Aytekin İ (2023). Robotik Sağımda Sürü Yönetimi. Turkish Journal of Agricultural Research 10(3), 361-371. https://doi.org/10.19159/tutad.1339586
10. Dado RG, Allen MS. (1995). Intake limitations, feeding behavior, and rumen function of cows challenged with rumen fill from dietary fiber or inert bulk. Journal of Dairy Science 78(1), 118-133.
11. Demir B, Öztürk İ (2010). Robotlu sağım sistemleri. Alinteri Journal of Agriculture Science 19(2), 21-27.
12. Erdman RA, Varner M (1995). Fixed yield responses to increased milking frequency. Journal of Dairy Science 78(5), 1199-1203.
13. Eyduran, E. (2020). ehaGoF: Calculates Goodness of Fit Statistics. R package version 0.1.0. https://CRAN.Rproject.org/package=ehaGoF
14. Gisi DD, DePeters EJ, Pelissier CL. (1986). Three times daily milking of cows in California dairy herds. Journal of Dairy Science 69(3), 863-868.
15. Grzesiak W, Zaborsk D (2012). Examples of the use of data mining methods in animal breeding. Data mining applications in engineering and medicine 303-324.
16. Hart KD, McBride BW, Duffield TF, DeVries TJ. (2013). Effect of milking frequency on the behavior and productivity of lactating dairy cows. Journal of Dairy Science 96(11), 6973-6985.
17. Himu HA, Raihan A (2024). An overview of precision livestock farming (PLF) technologies for digitalizing animal husbandry toward sustainability. Global Sustainability Research 3(3), 1-14.
18. IBM Corp. Released (2015). IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY: IBM Corp.
19. Isık AH, Alakus F, Eskicioğlu ÖC (2021). Hayvancılıkta robotik sistemler ve yapay zeka uygulamaları. Düzce Üniversitesi Bilim ve Teknoloji Dergisi 9(6), 370-382.
20. Jacobs JA, Siegford JM (2012). Invited review: The impact of automatic milking systems on dairy cow management, behavior, health, and welfare. Journal of Dairy Science 95(5), 2227-2247.
21. Ji B, Banhazi T, Phillips CJ, Wang C, Li B (2022). A machine learning framework to predict the next month's daily milk yield, milk composition and milking frequency for cows in a robotic dairy farm. Biosystems Engineering 216, 186-197. https://doi.org/10.1016/j.biosystemseng.2022.02.013
22. Johansen F, Buijs S, Arnott G (2025). Social rank and personality are associated with visit frequency in dairy cows learning to use an automatic milking system. Animal 19(3), 101446. https://doi.org/10.1016/j.animal.2025.101446
23. Johnston C, DeVries TJ. (2018). Associations of feeding behavior and milk production in dairy cows. Journal of Dairy Science 101(4), 3367-3373.
24. Kass GV (1980). An exploratory technique for investigating large quantities of categorical data. Journal of the Royal Statistical Society: Series C (Applied Statistics) 29(2), 119-127.
25. Kruip TAM, Morice H, Robert M, Ouweltjes W. (2002). Robotic milking and its effect on fertility and cell counts. Journal of Dairy Science 85(10), 2576-2581.
26. Maimon OZ, Rokach L (2014). Data mining with decision trees: theory and applications (Vol. 81). World Scientific.
27. Mathijs E (2004). Socio-economic aspects of automatic milking. In Automatic milking, a better understanding. Wageningen Academic. (pp. 46-55).
28. Morgan JN, Sonquist JA (1963a). Problems in the analysis of survey data, and a proposal. Journal of the American Statistical Association 58, 415–434.
29. Morgan JN, Sonquist JA (1963b). Some Results from a Non-Symmetrical Branching Processthat Looks for Interaction Effects. Proceedings of the Social Statistics Section, American Statistical Association 8, 40–53.
30. R Core Team (2020). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Retrieved from https://www.R-project.org/ (access date: 25.11.2024).
31. Rodenburg J (2017). Robotic milking: Technology, farm design, and effects on work flow. Journal of Dairy Science 100(9), 7729-7738. https://doi.org/jds.2016-11715
32. Simões Filho LM, Lopes MA, Brito SC, Rossi G, Conti L, Barbari M (2020). Robotic milking of dairy cows: a review. Semina: Ciências Agrárias 41(6), 2833-2850.
33. Tangirala S (2020). Evaluating the impact of Gini index and information gain on classification using decision tree classifier algorithm. International Journal of Advanced Computer Science and Applications 11(2), 612-619.
34. Topal M, Aksakal V, Bayram B, Yaganoglu AM (2010). An analysis of the factors affecting birth weight and actual milk yield in Swedish red cattle using regression tree analysis. The Journal of Animal and Plant Sciences 20(2), 63-6.9.
35. Tse C, Barkema HW, DeVries TJ, Rushen J, Pajor EA (2018). Impact of automatic milking systems on dairy cattle producers’ reports of milking labour management, milk production and milk quality. Animal 12(12), 2649-2656. https://doi.org/10.1017/S1751731118000654
36. Wei T, Simko VR (2021). Package “corrplot”: Visualization of a Correlation Matrix (Version 0.92). Package Corrplot for R Software.
37. Zaborski D, Ali M, Eyduran E, Grzesiak W, Tariq MM, Abbas F, Waheed A, Tirink C. (2019). Prediction of selected reproductive traits of indigenous Harnai sheep under the farm management system via various data mining algorithms. Pakistan Journal of Zoology 51(2), 421.
38. Zhang Y, Chi G, Zhang Z (2018). Decision tree for credit scoring and discovery of significant features: an empirical analysis based on Chinese microfinance for farmers. Filomat 32(5), 1513-1521.