As shown in the simulation-based process of Fig. 6, Task 1
(CAD/mesh of the part) is done in parallel with Task 2 which
consists of experimental parison extrusion tests to obtain
data for Task 3 (parison geometry and mesh generation for
model validation). Since Task 1 is of shorter duration than
Task 2, there is no increase in span time. In Task 4, simulation
of the blow moulding process using the mesh from Task
3 is carried out. The outcome of this task is a virtual part
having a weight and thickness distribution. The use of simulation
reduces not only the number of iterations required
to optimise the weight and thickness of the part, but also
the duration of the iterations. Tasks 8 through 11 are similar
to those of Fig. 5, tasks already described. The experimental
and simulation processes were modelled and simulated
using FirstSTEPTM with the appropriate data for duration,
rework estimates, delays, resource requirements, information
flows and constraints. Duration and rework estimates were
obtained from past experience for extrusion blow moulding
part development, and were normally distributed with regard
to their uncertainty. The performance measure of interest in
this study was cycle time. The results of the simulation runs
are presented in Table 5.