An interesting, discrete simulation model tends to contain numerous nontrivial interactions among its components. When writing such a simulation in an event-based language (in this discussion, Simula), one must "engineer" these interactions by event scheduling. Unfortunately, scheduling operations (such as Simula's ACTIVATE, CANCEL, and HOLD) are, in essence, low-level start and stop commands.
A methodology is presented for constructing a discrete simulation which is "less-procedural" than a traditional Simula-style simulation. By "less-procedural," we mean that our methodology emphasizes the cause and effect aspects of the model and "camouflages" the underlying control details inherent in such low-level process coordination. In our approach, the model is developed in terms of a production rule system (PS) whose working store contains facts about the state of the model and whose rules describe the behavior of the components (including delay information representing consumption of simulated time). The model is augmented by a notion of "roles" for components. Roles provide a simple behavior inheritance mechanism as well as a priority scheme which is used to resolve "incompatibilities" among component actions.
An "operational framework" is also described which elicits the appropriate behavior from the PS-based model. This framework employs a second PS whose working store contains "meta-facts" about the state of the model's actions and whose rules are "meta-rules" (synthesized from a static "dependency analysis" of the user-supplied rules). These meta-rules serve as constraints that coordinate the invocation, interruption, and resumption of the modeled components' actions. In addition, a "local agenda" is associated with each component to allow multiple component actions. The resulting framework can be mapped onto Simula's conventional process class and sequencing set (SQS). A working prototype has been implemented in Utah Portable Standard Lisp (PSL). The system is illustrated with an "office modeling" example.
Recommendations
Framework for Simulation of Hybrid Systems: Interoperation of Discrete Event and Continuous Simulators Using HLA/RTI
PADS '11: Proceedings of the 2011 IEEE Workshop on Principles of Advanced and Distributed SimulationA hybrid system is a combination of discrete event and continuous systems that act together to perform a function not possible with any one of the individual system types alone. A simulation model for the system consists of two sub-models, a continuous ...
Multi-fidelity modeling & simulation methodology for simulation speed up
SIGSIM PADS '14: Proceedings of the 2nd ACM SIGSIM Conference on Principles of Advanced Discrete SimulationM&S-based analysis has been performed for simulation experiments of all possible input combinations as a 'what-if' analysis causing the simulation to be extremely time-consuming. To resolve this problem, this paper proposes a multi-fidelity M&S ...