Analysis of Air Quality Parameters to Assess the Impact on Layers in Poultry Farms using Deep Learning

The food security has increased the agriculture production due to satisfying demand of ever-growing population. Due to this growth in population, the demand of protein also increased. A significant amount of population depends upon the chicken and egg to fulfil the demand of protein. The meat and egg production depends on the quality of poultry farming. The presence of air contaminants causes poor air quality within the poultry house which affects health of layers, production of eggs and workers in poultry farm. The proposed work uses data analysis approach and machine learning concept to automatize the process of air quality monitoring in poultry farms. A Convoluted Neural Network Long Short-Term Memory model, along with bidirectional Long Short-Term Memory model is proposed to improve the forecasting performance. This method predicts the Air Quality Index based on air quality parameters. The proposed approach is tested on poultry farm air quality dataset which is collected from different poultry farms. Finally, the obtained performance is compared with existing techniques in terms of RMSE, MAE, MAPE and correlation coefficient.