This paper introduces a comprehensive framework for automating process planning in machining, with a focus on capturing geometric transformations through voxel-based generative modeling. We have developed a multi-modal temporal dataset that is carefully structured around incremental changes in geometry, which are intricately linked to process parameters at each stage of production. By discretizing part transformations into small voxel units, our approach enables high-resolution tracking of shape modifications as material is progressively removed. This voxel-based representation, combined with detailed process parameters, facilitates precise spatial feature extraction. These extracted features serve as critical input for Auto-Encoder-based generative models designed to predict successive manufacturing steps. The curated dataset we have established will be used to train machine learning algorithms, enabling them to recognize the evolution of shape with respect to various process parameters and to predict future transformations needed to achieve the final geometry. This capability supports adaptive, data-driven decision-making, which is crucial for efficient and accurate process planning. Our framework is particularly valuable for Small and Medium Enterprises (SMEs), offering them a robust and scalable tool for dynamic process planning. By leveraging this data-driven approach, SMEs can optimize manufacturing sequences for complex parts, thus enhancing productivity and efficiency. This work ultimately provides a transformative, scalable methodology for the future of machining process planning.