The CMM (coordinate measuring machines) are able to perform dimensional inspections in workpieces with complex geometries, in a short time compared to conventional methods, however, errors on volume of CMM harm the performance of measurement. Faced with this, the purpose this investigation is to identify regions in the machine with large and small values of volumetric errors. The mapping of volumetric error is performed by optimization of objective function with SQP method. The objective function is defined by modeling errors of the CMM using a method of the homogeneous transformation, and, by calibration curves of individual errors. The optimization allowed to obtain the smallest value of volumetric error, 1.1796 μm, located near the linear encoder of the y axis. The mapping the volumetric error by optimization allows to know regions with minor harm the performance of measurement, therefore, it's possible to select regions of measurement to obtain reliable results.