<p>Scaling on the evaporator severely impacts heat transfer efficiency, reducing both production efficiency and product quality. Optimizing the cleaning schedule of the evaporator is crucial for improving heat transfer performance. This paper proposes a dynamic optimal decision-making strategy for determining the optimal cleaning timing. Firstly, a mechanistic model is established that accounts for the variation in material concentration over time, providing an accurate representation of the scaling process in sodium aluminate solution evaporation. Secondly, an infinite time domain cleaning optimization model is proposed, which considers the continuity of scaling between cycles. Lastly, a rolling optimization decision approach based on prediction is designed. The results demonstrate that the proposed method significantly outperforms traditional single-cycle optimization, scaling threshold-based methods, and time threshold-based cleaning methods in reducing cleaning frequency, improving heat transfer efficiency, and increasing net profit. </p>