SCSC2003 Abstract S3323
On-Line Simulation: Towards New Statistical Approaches
On-Line Simulation: Towards New Statistical Approaches
Submitting Author: Mr. Thomas Bessey
Abstract:
Complex systems such as flexible manufacturing systems or traffic systems typically evolve with alternating periods of transient and nearly steady-state behavior; such systems often show suboptimal performance. Thus, it is desirable to optimize the system's performance on-line by adjusting the system's parameters properly before a performance drop is to occur. To this end, the system's future evolution is assessed in advance repeatedly by means of on-line simulation. Since the system under control continues to evolve while the on-line simulation is performed, it is questionable whether classical statistical theory is still applicable to the issue of on-line simulation, possibly leading to unsuitable decision making.
We propose two new statistical approaches that are expected to at least relax the problem of unsuitable decision making that arises because of the system's ongoing evolution; along with this, we present some experimental results to give hints about the impact of our approaches on on-line control. To this end, we use a high-level Petri net formalism for modeling the system under control.
Back to SCSC2003 Abstracts