Abstract A problem when exploiting renewable energy sources, such as wind and solar radiation, is their fluctuating availability. In the presented work, the Heat Integration methodology for batch processes based on Time Slices has been extended to cover the integration of solar thermal energy, thus allowing for dealing with such variations. A procedure for identifying the number and durations of Time Slices for a problem featuring variable renewable energy supply has been formulated, and developed for solar energy utilisation. The main procedural steps involve partitioning of the measured/forecasted heat availability profile using a large number of candidate time boundaries, and then approximating it by a piecewise-constant profile using high-precision. The approximation profile is obtained by subjecting the candidate superset of time-boundaries to MILP optimisation, thus minimising integral inaccuracy compared to the forecasted availability profile. The Time Slice definitions are completed by approximating the heat loads within the Time Slices. The integration of solar thermal energy can be performed for the specified Time Slice, after the optimal number of Time Slices with approximated constant load has been selected. Using heat storage, the heat can be transferred between Time Slices.