A significant part of the activities in a manufacturing system involve assembly tasks. Nowadays, these tasks are object of automation due to the market increasing demand for quality, productivity and variety of the products. Consequently, the automation of assembly systems should consider flexibility to face product diversification, functionalities, delivery times, and volumes involved. However, these systems are vulnerable to faults due to the characteristic of their mechanism and the complex interaction among their control devices. In this context, the present work is focused on the modeling design of flexible assembly systems control, including the occurrence of faults. The proposed method structures a sequence of steps for the models construction of assembly processes and their fault detection, based on the theory of discrete events systems and Petri net. This work use in special, production flow schema/mark flow graph (PFS/MFG) technique to describe and model the flexible assembly systems control through a rational and systematic procedure, as well as, the processes data record based on quantitative techniques for fault detection. This approach is applied to a flexible assembly systems installed and in operation to compare the effectiveness of the developed procedure.