<p>Due to increasing demand for livestock products, livestock production systems have become larger and more intensive, resulting in increased manure generation and stronger odour emissions. Livestock odour emission factors are influenced by multiple environmental and operational factors, including farm characteristics, meteorological conditions, ventilation systems, and gas concentrations. However, conventional approaches generally apply fixed odour emission factors that may not adequately represent changing environmental conditions. This study developed an integrated livestock odour dispersion forecasting system by combining machine learning-based odour emission prediction with an air dispersion model. Machine learning regression models were developed using farm operational information, meteorological conditions, and gas concentrations to predict livestock odour emission factors. Among the evaluated models, the Random Forest (RF) model showed the highest prediction performance, with an <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({R}^{2}\)</EquationSource> </InlineEquation> of 0.96, RMSE of 6.8 OU·s⁻¹· animal⁻¹, and MAPE of 0.12%. The selected RF model was integrated with the AERMOD air dispersion model to develop a web-based livestock odour dispersion forecasting system (FLOD). The developed livestock odour dispersion forecasting system utilized real-time and forecasted meteorological data together with farm operational information to dynamically estimate odour dispersion around livestock facilities. The results demonstrated the potential of integrating machine learning-based odour emission prediction with air dispersion modelling for operational livestock odour assessment and management.</p>