In a semiconductor manufacturing system, particular human operations may violate desired requirements and lead to destructive failure. For such human-in-the-loop systems, this paper proposes a supervisory framework which guarantees that manual operations meet required specifications so as to prevent human errors in operation using Petri nets. Moreover, a modular technique with an intersection mechanism is proposed in order to cope with the state-space explosion problem of large-scale systems. A rapid thermal process in semiconductor manufacturing systems is provided to show the practicability of the proposed approach. Note to Practitioners - This work was motivated by the requirement of remote monitoring and supervision of semiconductor manufacturing systems. In such human-in-the-loop and large-scale systems, certain human operations may violate desired safety requirements and result in catastrophic failure. Moreover, the overall complexity of most existing Petri net modeling and analysis approaches significantly increases with the size of the considered systems. This paper suggests a modular supervisory scheme for modeling and synthesis of supervisory agents using Petri nets. The proposed method contributes a promising tool for preventing abnormal operations from being carried out to semiconductor manufacturing systems which, if appropriately modified, may be also applied to other types of discrete event systems.