Precise knowledge of the scale mass formed is essential for the production of high quality steel. In this context several different models have been developed to represent the formation of the scale layer, all of these based on a purely parabolic approach. In the present work computational fluid dynamics (CFD) is used to characterize gas phase combustion and steel heating in a walking beam type reheating furnace. The developed model is based on two separate simulations. The great advantage here is that the elaborate combustion calculation is performed in a stationary simulation, while the transient heating is considered in a comparatively small domain minimizing computational effort. This allows the usage of a scale formation model without an excessive increase in computing time, with the results that this method is numerically highly efficient. In this work a user-defined function (UDF) has been applied to characterize scale formation. The model is able to calculate local scale mass based on steel temperature, time as well as the associated species concentration in the surrounding atmosphere. In addition, the model is able to depict the insulating effect due to the low thermal conductivity of the scale. The influence turned out to be minor, since the heating time is low compared to other similar furnaces. This has been evaluated by comparing the results to the model neglecting the insulating effect. The results are further compared with a analytical model based on a parabolic approach. Moreover, measurements on the real furnace have been performed. The weight of several tubes before and after heating has been recorded, as indicators of the scale mass formed. Measurements in real furnaces represent a major challenge, and the results show a very close agreement with the numerical results we have achieved. (C) 2018 Elsevier Ltd. All rights reserved.