Physics-based machining models typically require significant computational resources and time which makes them impractical for quick process optimization calculations in an industrial shop-floor setting. In this paper, we demonstrate the use of metamodeling to help overcome the inefficiency of traditional engineering modeling tools. Metamodels are surrogate models of an existing model that provide an approximate relationship between a set of input process variables and output response variables. Using 2D orthogonal machining as a case study, we utilize the Response Surface Methodology to develop surrogate models which provide approximate relationships between machining variables such as cutting tool speed, depth of cut, rake angle, and tool and workpiece materials, and response variables such as maximum residual stress, residual stress transition depth, surface hardness, tool wear, and wear depth. The Kriging interpolation method is used to generate response surface maps. Finally, we develop the idea of a mobile application which can be easily deployed in an industrial setting to efficiently use such metamodels for physics-based decision making during the initial manufacturing process development phase.