In this paper, a neural network approach to the pork farm odour using single-component analysis with the consideration of other relevant factors, such as measurement location is proposed. Neural network models and statistical models for pork farm odour have been developed and compared for single-component models to determine which method produces superior results. Non-linear statistical and neural network models were developed and compared for single-component pork farm odour analysis using NH3 and H2S as odour indicators.  The first set of models considered only a single-odour indicator and no other factors. In this case, the odour intensity predictions from the neural networks were generally better than the predictions produced by the non-linear statistical models.  The second set of models considered asingl e-odour indicator (NH3 or H2S) and the odour source (pig building or pig manure storage). In this case, both of the developed neural network models N5 and N6 performed better than their corresponding neural network models (N1–N4) for every performance measure considered in this paper. In general, the use of neural networks to model the pork farm odour yields more accurate and precise odour intensity predictions than the statistical models. The measurement location for the pork farm odour was considered in several model comparisons. The neural network models significantly outperformed the statistical models in this comparison because the statistical models are not able to consider the measurement location. This indicates that measurement location is a relevant factor for modelling pork farm odour. This also demonstrates that factors other than odour components should be considered during modelling.