The demand for cement has significantly increased, growing by 8% in the year 2022 and by a further 12% in 2023. It is highly anticipated that this trend will continue, and it will result in significant growth by 2030. However, cement production is highly energy-intensive, with 70 to 80% of the total energy consumed during the clinker formation, which is the main cement production process. Minimising energy losses requires a radical approach that includes optimising the performance of the kilns and significantly improving their energy efficiency. One of the most efficient approaches to optimise the performance of the kilns and reduce energy losses is by integrating process re-engineering models, which leverage process data analytics, machine learning, and computational methods. This study employed a model-based integration approach to improve energy efficiency during clinker formation. Energy consumption data were collected from two semi-automated cement production plants. The data were analysed using a regression model in Minitab (Minitab 21.1.0) statistical software. The analysis resulted in a linear energy consumption equation that links energy consumption to both production and energy loss. Dynamic simulations and modelling using Simulink in MATLAB were performed based on a proportional–integral–derivative (PID)-controlled system. The dynamic behaviour of the model was evaluated using data from Plant A and validated with data from Plant B. The energy efficiency equation was established as a mathematical model that explains energy improvements based on incorporating parameters for the cement kiln system and disturbances from the environment.