The heat removal rate from the solidifying molten metal is one of the most important issues in casting processes. The analysis of the inverse heat conduction problem (IHCP) with an assumption of one‐dimensional heat flow is a typical approach to clarify the heat transfer characteristics between the solidifying metal and the mould, and the authors have also developed a single‐belt strip casting simulator applying this technique. In this study the effects of the mould geometry and the temperature measurement on the estimation accuracy of surface temperature and interfacial heat flux in this method were examined through numerical experiments. The mould profile of the simulator where a side dam is integrated generates a multi‐dimensional heat flow and causes a non‐uniform heat flux along the metal‐mould interface. The heat flow in the central part, however, is unidirectional in the initial period and the thermal status at the interface can well be reproduced by the IHCP analysis. Furthermore, the disturbances in the temperature measurement, for example the interfacial thermal resistance between the inserted thermocouples and the casting mould, directly cause estimation errors and should be removed. These results indicate that the single‐belt strip casting simulator developed by the authors is capable to measure the interfacial heat flux and the surface temperature in the casting process although the measuring period and the location are limited. They also suggest the methods how to imporve the measurement accuracy for the various casting simulation techniques.