Characteristics of pharmaceutical products are influenced by a number of factors relating to the formulations and manufacturing conditions. To optimize these factors, the formulator must consider a wide range of characteristics such as the disintegration time, dissolution rate, hardness and stability. In most cases, the relationships between the formulation factors, process variables and output characteristics are rather complicated, and the effects of the causal factors are substantially changed by the changing physicochemical properties of the active pharmaceutical ingredients. “Pharmacometrics” can be defined as chemometrical technology for analyzing complex phenomena often found in pharmaceutical research and product development. Pharmacometrical analysis utilizes notable statistical and mathematical methods such as the experimental design (DOE), response surface method (RSM), multivariate analysis and artificial intelligence. Numerical simulations such as the finite element method (FEM) and discrete element method may also be included in the broad sense of pharmacometrics. In this review, the basic concept of DOE and RSM is introduced first, and in particular detail, the nonlinear RSM incorporating a multivariate spline interpolation. The tablet database composed of well-organized formulations is exemplified as a typical case of multivariate analysis. Kohonen's self-organizing map is proposed as a powerful tool for visualizing the latent structure underlying the tablet database. Finally, several applications of FEM are introduced for simulating residual stress distributions in compressed tablets.