The chamber volumes are very important for some mechanical products. For instance, the volume variations of engine cylinder head combustion chambers directly affect the critical functions (e.g. compression ratio) of an engine. The interior surfaces of the chambers are usually not being machined after casting processes due to high machining cost. Traditional titration methods are only applied off line to evaluate the variations of the chamber volumes since they are considerably time-consuming. Therefore, it is difficult to on line control the volume variations of multiple chambers in machining processes. With the development of new high definition metrology (HDM) technologies, millions of coordinate points of the interior surfaces of the chambers can be on line obtained, and thus great opportunities are provided for on-line controlling volume variations of multiple chambers of a workpiece. However, there are some critical problems urgently need to be solved, such as datum transformation of high-density points, precise volume calculation of multiple chambers, and minimizing the volume difference of any two ones of all chambers. This paper presents a novel systemic approach for on-line minimizing the volume difference of multiple chambers of a workpiece based on HDM. A model for obtaining an optimized machining parameter for depth of chambers is explored to minimize the volume difference of any two ones of all chambers. The results of a case study show that the proposed approach can minimize the volume difference of four combustion chambers of a cylinder head in machining processes.