נגישות
menu      
חיפוש מתקדם
תחביר
חפש...
הספר "אוצר וולקני"
אודות
תנאי שימוש
ניהול
קהילה:
אסיף מאגר המחקר החקלאי
פותח על ידי קלירמאש פתרונות בע"מ -
Predicting calving time of dairy cows by behaviour sensor
Year:
2011
Authors :
אנטלר, אהרון
;
.
הלחמי, אילן
;
.
מלץ, אפרים
;
.
Volume :
Co-Authors:
Maltz, E., ARO, Volcani Center, Institute of Agricultural Engineering, P.O.B. 6, Bet Dagan, Israel
Medini, N., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Bercovitch, A., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Parmet, Y., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Halachmi, I., ARO, Volcani Center, Institute of Agricultural Engineering, P.O.B. 6, Bet Dagan, Israel
Antler, A., ARO, Volcani Center, Institute of Agricultural Engineering, P.O.B. 6, Bet Dagan, Israel
Edan, Y., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Facilitators :
From page:
464
To page:
475
(
Total pages:
12
)
Abstract:
Predicting approaching calving enables proper supervision and work planning in the dairy. Behavior parameters monitored automatically for each cow (Afiact Plus®), were used to construct a model to predict calving time as a categorical variable. Visual analysis indicated significant changes in most of the parameters on the day before calving. The model's performance measures were maximum true positives (calving took place within 24 hours) and minimum false alarms. The best results were achieved by the Discriminant Function using as the minimizing variance transformation the ratio between successive days difference to the standard deviation of the average of the previous three days. This yielded 80.95% true positives and 22.80% false alarm. Extending prediction to 48 hours increased accuracy to 90.48% true positives and 15.60% false alarms indicating that automatically recorded behaviour parameters can be used to predict approaching calving.
Note:
Related Files :
Agriculture
Behaviour
Calving
Dairy cow
Discriminant functions
Forecasting
Lying time
Performance measure
עוד תגיות
תוכן קשור
More details
DOI :
Article number:
0
Affiliations:
Database:
סקופוס
Publication Type:
מאמר מתוך כינוס
;
.
Language:
אנגלית
Editors' remarks:
ID:
27285
Last updated date:
02/03/2022 17:27
Creation date:
17/04/2018 00:29
Scientific Publication
Predicting calving time of dairy cows by behaviour sensor
Maltz, E., ARO, Volcani Center, Institute of Agricultural Engineering, P.O.B. 6, Bet Dagan, Israel
Medini, N., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Bercovitch, A., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Parmet, Y., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Halachmi, I., ARO, Volcani Center, Institute of Agricultural Engineering, P.O.B. 6, Bet Dagan, Israel
Antler, A., ARO, Volcani Center, Institute of Agricultural Engineering, P.O.B. 6, Bet Dagan, Israel
Edan, Y., Ben-Gurion University of the Negev, Faculty of Engineering Sciences, Department of Industrial Engineering and Management, P.O.B. 653, Beer Sheva, Israel
Predicting calving time of dairy cows by behaviour sensor
Predicting approaching calving enables proper supervision and work planning in the dairy. Behavior parameters monitored automatically for each cow (Afiact Plus®), were used to construct a model to predict calving time as a categorical variable. Visual analysis indicated significant changes in most of the parameters on the day before calving. The model's performance measures were maximum true positives (calving took place within 24 hours) and minimum false alarms. The best results were achieved by the Discriminant Function using as the minimizing variance transformation the ratio between successive days difference to the standard deviation of the average of the previous three days. This yielded 80.95% true positives and 22.80% false alarm. Extending prediction to 48 hours increased accuracy to 90.48% true positives and 15.60% false alarms indicating that automatically recorded behaviour parameters can be used to predict approaching calving.
Scientific Publication
You may also be interested in