SIMULATION

The paper presents a review in the field of knowledge based simulation systems (KBSS). Development and application of such systems allow the user to perform simulation studies without special training in programming languages, simulation theory, and other skills related to simulation project realization. KBSS architecture, programming paradigms and development tools are considered. Problem areas and characteristics of the main applications of KBSS are described. More than 50 examples of research and commercial systems are given in the tables.

Quality assurance techniques which may be used in KBSS are considered. They support tools for model reduction and parallel processing techniques. The perspectives for KBSS development are discussed. Among them are support of both on-line digital control, simulation modelling and data base management, development of parallel and distributed systems, fuzzy simulation environment development, combination of the expert system capabilities with potential for adaptation by artificial neural nets. Bibliography includes 168 references.

VISTA (VISual Translucent Algorithm) is a fast, 4-D cloudscape generator that includes simulation of haze and cloud- related visual phenomena such as cloud shadows, rainbows, and solar arcs. VISTA uses a meteorological sounding as input and produces cloud scenes using a backward ray tracing technique. The ray is checked for the first intersection of either 1) an opaque surface (thick cloud or surface), 2) a translucent layer, or 3) an optical refraction layer. Depending on the nature of this intersection, the specular reflection is calculated after taking into account direct illumination, if any. Placement of clouds depends on location of potential cloud layers, cloud domain characteristics, and gravity waves, while small (of order 1 meter) cloud structure is stochastically generated by a sawtooth generator. Background land surfaces can be deterministically (either real or hypothetical) input or can be stochastically generated.

Traditional simulation techniques perform poorly when estimating performance measures based on rare events. One solution to this problem is the use of importance sampling. However, two problems that have limited the use of importance sampling are the lack of a formal framework for specifying importance sampling strategies, and the fact that in most cases the simulations must be hand-coded — a very time-consuming process. This paper presents a software tool that facilitates experimentation with importance sampling by addressing these two problems. First, the tool is based on a flexible framework for specifying importance sampling simulations in terms of stochastic activity networks. Second, once specified, the importance sampling simulation program is automatically generated by the tool, freeing the researcher to focus on the modeling problem. The effectiveness of the software is demonstrated through the solution of a machine-repairman model with Weibull distributed failure times and a delayed group repair policy. Orders of magnitude reduction in the CPU time required to obtain a specified relative accuracy were achieved.

The paper presents a simulation modeling environment developed for sawmill design and analysis in the forest products industry. The design facilitates flexibility in modeling different sawmill configurations and production scenarios. The system represents a library of objects developed in an object-oriented framework. These include structures required to develop simulation models, to execute discrete event simulation, and for system-user interface.