A methodology for quantitative assessment of the features and capabilities of software frameworks for model composition

Abstract

A software framework is an infrastructure or architecture intended to enable the integration and interoperation of a set of software components. A specialized type of software frameworks are those specifically designed to support the composition of models or other components within a simulation system. Such frameworks are intended to simplify the process of assembling a complex model or simulation system from simpler component models as well as to promote the reuse of the component models. Several different types of software frameworks for model composition have been designed and implemented; the various framework types have different component types, processes for composing models, and intended applications. The different framework types and the implemented examples of them vary widely in terms of features and capabilities. Comparing alternative frameworks so as to assess their likely utility for a specific application has heretofore been largely subjective and qualitative, and consequently of uncertain reliability. To address this issue an assessment methodology specifically designed for comparing model composition frameworks is developed and explained. The methodology employs a quantitative metric based on a set of well-defined criteria relating to the features and capabilities of simulation frameworks and is intended to be objective and quantitative. The methodology is illustrated and demonstrated by applying it to a set of existing model composition frameworks. The assessment results suggest that using the methodology can, at a minimum, improve the objectivity and reliability of framework selection decisions. The assessment criteria may also be useful as guidelines when designing and developing a framework.