Models based on regression analysis for predicting the operational efficiency of high-pressure roller crushers are considered. A systematized and structure

We discuss the development of a comprehensive data analysis method which allows increasing the accuracy of predicting the performance of roller mill of processing plant (RP) mining and processing plant (GOK) when you change the properties of incoming raw materials for processing. The described method includes primary data processing, determination by statistical analysis methods and data mining algorithms of the most significant factors affecting the resulting parameter, development of mathematical models based on correlation regression and factor analysis, analysis and confirmation of the quality of forecasting by a neural network. The systematization and coordination of production data was carried out, the generated database was processed using statistical and intellectual analysis methods, the physical and chemical parameters of the input raw materials and processed ore that have the greatest impact on the mill productivity were determined, mathematical models were formed to determine the expected productivity, their quality and applicability limits were evaluated, the error in predicting the mill productivity for various mineralogical compositions of the processed ore was determined. The significance of the parameters used in the models is checked by the algorithms of intelligent analysis. The verification of the used models for predicting mill performance by an artificial neural network is carried out by comparing the series of predicted values of the resulting factor obtained by different models

The present study provides a systematic benchmarking of nine state-of-the-art object detection models adapted to a specialized animal dataset. The main objective was to evaluate their performance in terms of accuracy and inference speed in the context of real-time and commercial applications within the agro-industrial sector. The method-ology involved aggregating and thoroughly cleaning several cattle image datasets, fol-lowed by fine-tuning models with diverse architectures — two-stage, one-stage, and transformer-based. Performance evaluation was conducted on a unified hardware platform equipped with an NVIDIA RTX 4090 GPU with 24 GB of VRAM, which is critical for the objec-tive assessment of speed-related metrics. Key results show that the D-FINE and Co-DETR models demonstrated the highest ac-curacy (AP@[IoU=0.50:0.95] of 0.872 and 0.851, respectively), while RTMDet and YOLOv11 proved to be the fastest (15.81 ms/image and 19.14 ms/image in reproducible mode). A general trend was observed toward significant improvement in average pre-cision (AP) metrics on the specialized dataset compared to their known results on gen-eral-purpose datasets such as COCO, while relative inference speed rankings remained largely consistent.

This study benchmarks nine state-of-the-art object detection models on a specialized cattle dataset to assess accuracy and inference speed for real-time agricultural applications. Using a unified protocol without model-specific augmentations, and evaluating all detectors on identical RTX 4090 hardware, we provide a fair architectural comparison of two-stage, one-stage, and transformer-based models. D_FINE_L and Co_DETR_R_50 achieved the highest accuracy (AP@[0.50:0.95] = 0.872 and 0.851), while RTMDet and YOLOv11_L were the fastest (15.81 and 19.14 ms/image). All models showed substantial accuracy gains on the domain dataset compared to COCO, while maintaining consistent relative speed rankings.