Rapid and customized part realization in all industrial sectors imposes stringent demands on part attributes, e.g., mechanical properties, microstructure, surface finish, geometry, etc. However, part attributes can very rarely be directly measured and/or controlled in the production process. Instead, measurements are taken of accessible and measurable primary process responses that are known to influence the part's attributes. These primary process responses are then controlled through the manipulation of a set of controllable process parameters. This widely used strategy is based on the assumption that the proper control of the primary process responses will implicitly yield the desired part attributes. The current work aims to replace this implicit assumption by a model-based explicit evaluation of the part's attributes that uses newly established process models, available measurements of process responses and historical data from a data base that is continuously updated. In effect, this approach implies a direct instead of an implicit control of the part's desired attributes and, as such, also moves a step closer to rapid part certification. The research will establish the scientific and technological foundation for a manufacturing platform in a distributed network that seamlessly and efficiently integrates physical processes and numerical simulations in a fast predictive framework. The platform is envisioned as a multi-loop simulation and control environment consisting of four control loops running at different time scales. Two of the control loops, similar in structure to conventional controllers, act at the hardware-level and are devoted to the physical control of the relevant process variables while the other two are devoted to the software-level model-based evaluation of the desired part attributes. The latter two instruct the hardware-level controllers on required changes in their behavior that are necessary to reach the desired part attributes. To enable the integration, a voxel-based geometric model powered by an underlying data structure capable of dynamically generating analysis information, storing experimental information, and encoding the final part attributes obtained from the simulation and measured results will be established. This geometrical representation is well-suited to the use of general purpose graphics processing units (GPGPU) for fast computation of the process models that determine the physical process responses and attributes in arbitrary regions of a part. The researched framework will be validated using the state-of-the art open-architecture Directed Energy Deposition machine at Northwestern equipped with networked real-time sensing and control.