This paper develops an analytical approach to quantifying manufacturing system flexibility (MSF). Measures of flexibility, namely, producibility, processivity, transferability and introducibility are developed based on a state-transition formalism, the reachability graph. The uncertainty in system operating conditions is specified in terms of certain disturbing factors, external and internal to the system, which impinge on system performance. The contribution of each factor to the value of a given type of flexibility is estimated using the analytic hierarchy process (AHP). The individual flexibility measures are thus combined to give a measure of MSF under a given operating environment. A Petri net model is used to construct the reachability graphs used in flexibility measurement. Uses of the proposed measures at the design and operational analysis stages of FMS are presented