A novel Neural Network Based Controller (NNBC) has been developed following a comprehensive set of experiments carried out on a pilot-scale stoker test facility. The subsequent modification and evaluation of the neural control system on an industrial chain grate stoker fired boiler is described in this paper. The NNBC simulated the actions of an expert boiler operator, by providing initial estimates of the near optimum settings required for the coal feed and airflow rates as well as appropriate staging of these variables during load-following conditions. When quasi-steady-state combustion conditions were attained after a load change, the combustion air Row rate was fine tuned to minimise the excess air level. Test results for the controller have demonstrated that improved transient and steady-state combustion conditions were achieved without adversely affecting pollutant emissions or the integrity of the stoker and boiler system. The prototype NNBC thus has potential to provide both stoker manufacturers and users with a means to control pollutant emissions as well as improving the combustion efficiency of this type of coal-firing equipment.