Comparison of simulation modeling techniques that use preemption to capture design uncertainty

This paper describes a process, implemented using two simulation engines that adopt, respectively, the event scheduling paradigm and the activity scanning paradigm. The process being modeled is design development in an unpredictable environment. Unpredictability means that criteria are prone to change during design, thereby interrupting ongoing work and causing design iteration. Probability density curves, input to the simulation, capture uncertainties regarding design criteria during the development of R & D semiconductor fabrication facilities. The simulation of process changes calls for preempting tasks or events, and scheduling new tasks or events. The implementations in alternative modeling paradigms illustrates the use of a top-down vs. a bottom-up approach in process modeling. The two engines that were used, SIGMA and STROBOSCOPE, both, are programmable so that the model could be implemented without difficulty in either one.