GESCONF: real-time evaluation of the zootechnical performance of feedlot beef cattle.
beacon, esp32, flutter, nutrition, modeling.
The use of computational modeling is an alternative to assist beef cattle farmers in decision making, especially in feedlot systems. The development of models to predict the growth curve of confined animals, based on the behavioral monitoring of cattle, can reduce the time and risks of adopting a new diet and allow estimating the result of zootechnical performance, as well as predicting the time of adoption ideal exit for slaughter. In this sense, the objective of the dissertation was to develop an automated system for real-time evaluation of the zootechnical performance of beef cattle in feedlots. The project is being developed at the Federal Institute Catarinense, Campus Araquari, Santa Catarina. On August 15, 2020, twelve male castrated with an average weight of 342 kg, crossbred European breeds were confined in a closed shed. All were weighed at the entrance and three of them were identified in the ear with a beacon before entering the feedlot. Feeding is carried out twice a day, and along the process, they received two different diets (fattening and finishing), in addition to the transition diets. One of the cattle remains in an individual stall, where the feed is weighed before delivery as well as the leftovers present in the trough, every 15 days the animal in the individual stall is weighed and changed at random. Next to the trough of the individual stall, an esp32 module was installed that will identify the animal and quantify the frequency with which it goes to the trough and the time it spends eating. This trough was placed on a bench-balance with bluetooth, allowing to monitor the consumption of the diet. The individual drinker was fixed on an electronic platform balance with bluetooth allowing to monitor the weight gain. Both balances were synchronized to a cell phone that sends the data to the store on the Firebase Cloud platform. From this information, together with the bromatological analysis of the diets provided, the body development model will be fed providing an estimate of the animal's weight, as well as its slaughter point. This information will be generated from the Flutter framework. Blood samples from the animals are being performed monthly to determine the metabolic parameters. Two intermediate weighings, in addition to those of entry and exit in the feedlot, were carried out. At the end of the experimental period (slaughter of the animals), the weights measured will be correlated with the weights estimated by the model with the aid of the statistical program R, whose significance level will be 5%.