The technology develops an algorithm to estimate the average weight of broilers in poultry houses every hour. In addition, using the prediction results and historical broiler carcass distribution information to predict the average weight and slaughtering specification distribution of each farm for the next 7 days. The technology enables agribusiness to accurately schedule the appropriate time for each farm to meet the market＇s specifications and reduce losses and waste. The technology has been validated in the broiler industry, achieving an accuracy of 98％ in weight prediction and an error rate of 6％ in specification prediction.