The ash fusion temperature (AFT) profile of a coal is one of the parameters currently widely used in coal marketing and utilization to assess coal ash fusibility and melting characteristics, and to predict the melting behaviour of the coal ash in coal conversion processes. However, the flow temperature of a specific coal source alone does not provide enough insight into the slag formation behaviour at temperatures below the flow temperature. It has been demonstrated that ash flow temperature can be correlated with FACT (TM) equilibrium calculations, and that such equilibrium calculations provide useful information regarding the phase transitions that take place well below the final ash flow temperature as indicated by a standard AFT analysis. The overall objective of this study is to use the properties of different types of mineral matter and to simulate the effect on slag-liquid formation and ash flow temperature for individual mineral types as well as blends. A thermo-equilibrium simulation approach based on FACTSAGE (TM) was applied. The slag-liquid flow temperature simulations for coal and individual mineral types compared favourably with the actual measured ash flow temperature and are within the experimental error of an AFT analysis (+/- 30 degrees C). The slag formation of coal showed a more gradual increase in the amount of slag formed over a temperature compared to specific individual mineral types. This can be explained by the ash composition of coal which has a more even distribution of minerals compared to mineral types that have a significantly high SiO2 or CaO content respectively. From this result it is also confirmed that flow temperature properties are non-additive and cannot be calculated from a weighted average principle. It was shown by FACTSAGE (TM) modelling that the addition of a discard blend to this specific coal source will have no significant effect on the flow temperature of the blend. The FACTSAGE (TM) modelling showed that the amount of slag-liquid of the ash that was studied is highly dependent on the temperature to which the ash is exposed. For example, the amount of slag-liquid can double by only increasing the operating temperature of a process within a small window of as little as 100 degrees C. This could explain why coal conversion processes such as combustion, and entrained flow gasification, can experience periods of unstable operation despite an apparent constant feed coal quality being observed with very little variation in ash flow temperature. The important parameter is thus not the actual flow temperature, but rather the amount of slag formation at a specific operating temperature. The inability of the standard AFT analysis to reflect changes in ash melting behaviour with relatively small variations in ash composition was highlighted in this study. Even though variation in mineral matter content of the coal blend may have limited or no effect on the actual ash flow temperature, it can significantly change the amount of slag present in the ash bed. The variation in operating temperature can also result in a significant change in slag formation (positive or negative) which is not always reflected by the standard AFT analysis method. It is concluded that FACTSAGE (TM) thermodynamic modelling of ash melting behaviour provides a method to address the shortcomings of the standard AFT determination, which can add significant value and insight towards the improvement of the operational stability of coal conversion processes such as combustion and gasification. (C) 2013 Published by Elsevier Ltd.