Abstract: State space model, which is based on the state space method in control theory, is applied to reveal the propagation, transformation and accumulation of part variations in multista-tion assembly processes (MAP). Taking the fixture variations for input vectors and part variations for state vectors, a state function is built to represent the relationships between fixture variations and part variations, also the reorientation variations when assemblies are transformed between different stations. Taking the measurement values of parts for output vectors, an output function is built to represent the number and position of the sensors. Finally, a real MAP case in automotive assembly line is given to illustrate the method presented.

Key words: Multistation assembly processes, State space model, Variations stream