In this paper, multivariable systems theory is used to synthesize control policies for a manufacturing system which is modelled as a linear discrete-time multivariable dynamical system. The state variables of the system are the rates-of-flow of parts at the various work stations, the buffer inventories, anil the finished product inventories. The control inputs are the man-hours per week required for the various processes involved in the manufacturing system, anil the disturbance inputs are the demands for semi-finished and finished parts. The behaviour of the controlled manufacturing system is illustrated by the presentation of the results of computer simulation studies showing typical transient and steady-state response characteristics,