Enhanced Prediction of Broiler Shipment Weight Using Vision-Assisted Load Cell Analysis

Accurate prediction of broiler shipment weight is essential for optimizing production planning and meeting market demand. Previous studies have estimated representative daily weight values from load cell data using K-means clustering and kernel density estimation (KDE), and have applied forecasting models such as Prophet, ARIMA, and Gompertz. Among these, the combination of K-means and Prophet demonstrated the best performance. In this study, we propose an enhanced method that integrates computer vision with load cell measurements. A YOLOv8-based object detection algorithm is employed to count the number of broilers on the load cell using real-time images captured by a camera. The average weight per broiler is estimated by dividing the total measured weight by the number of detected chickens. Based on the previously established forecasting framework, these more accurate values are fed into the Prophet model to predict shipment weight. Experimental results show that, compared with earlier methods, the proposed approach improves prediction accuracy by approximately 2.82%, enabling a better understanding of broiler growth patterns and providing more reliable support for shipment scheduling in smart poultry farms.