The flotation cells used to upgrade ultrafine metallurgical coal (-250 µm) are usually poorly controlled leading to unnecessary loss of combustibles from the flotation system. This is mainly because coal flotation cells are typically controlled for process stability rather than optimisation of flotation performance. This is especially true for a flotation device commonly used in coal applications in Australia – the Jameson Cell. This then provides an opportunity to improve the flotation performance by shifting from a stabilising control strategy to a proactive optimising control strategy. For an optimising control strategy, the control system will require a means of relating and then measuring in real-time, the effect of changing operating conditions on the coal flotation performance. Hence, the major aim of this work is to propose a possible model-informed process control strategy that could be used for real-time performance optimisation. In this study, industrial site work was conducted on an industrial-scaled Jameson Cell at a Coal Handling and Preparation Plant (CHPP) in Central Queensland, Australia where superficial gas velocity was controlled to six different levels. At the same time, other process variables such as froth depth and flotation feed properties were measured. The empirical models developed from this study suggested that combustible recovery is a function of superficial gas velocity, froth depth, feed solid content and feed ash content and that a peak in combustible recovery existed as a function of superficial gas velocity. This led to the development of a model-informed control strategy that aimed to maximise the combustible recovery by determining and then operating the cell at the superficial gas velocity at which the peak occurs. This peak superficial gas velocity was found to be only a function of feed solid content. Based on these findings it would seem possible to operate a Jameson cell at peak performance by online measurement of feed solid content, prediction of the optimum superficial gas velocity and stabilising control to maintain operation at that superficial gas velocity set point. The development of this model_x0002_informed control strategy can serve as a basis for the development of a more sophisticated model_x0002_predictive control (MPC) system. However, more work will need to be done to improve the predictivity of the combustible recovery model by measuring and modelling performance over a wider range of operating conditions.