PM10 emission forecasting using artificial neural networks and genetic algorithm input variable optimization

► Neural network (ANN) modeling of annual PM10 emissions at a national level ► Sustainability and economical/industrial parameters are used as model inputs. ► The selection of inputs was based on smoothing factor (ISF) calculated by GA. ► The ANN model provides much better results in comparison with conventional models. ► Up to two years forecast with the ANN model can be made successfully and accurately.