What Are the Four Types of Models?

The different types of systems and their various aspects are modelled through 4 primary types of models and classifications of those models. It is crucial the right type of model is chosen to effectively model the purpose and scope.

“A model can be [a] physical, mathematical, or otherwise logical representation of a system” as defined by the United States “Department of Defense Modeling and Simulation (M&S) Glossary”. 

  1. Physical Models

A physical model is exactly that. A concrete representation, separated from the abstract mathematical, and logical models. Abstract models can be descriptive or logical in order to be representative of the purpose and scope but are actual physical representations.

  1. System/Schematic Models

Some system models are often a mix of both analytic and descriptive, they will also cover multiple integrated domains to ensure a complete and functional system. These constructs will provide multiple views to support the requirements, design, analysis, planning and verificatory components.

In the Mathematical Theory of Systems Engineering: The Elements (Wymore 1967), we learn from Wayne Wymore who formally defined a system model as “a rigorous mathematical framework for designing systems in a model-based context.” A summary of his work can be found in A Survey of Model-Based Systems Engineering (MBSE) Methodologies.

Wayne Wymore formally defined a system model in Mathematical Theory of Systems Engineering: The Elements (Wymore 1967). As “a rigorous mathematical framework for designing systems in a model-based context.” A summary of his work can be found in A Survey of Model-Based Systems Engineering (MBSE) Methodologies.

  1. Verbal/Descriptive models

Descriptive models use language to describe the relationships, such as a parts tree, the functions of its components, or test cases. Often the physical and functional architecture of the system is described in great detail to better demonstrate the scope and purpose.

  1. Mathematical/Analytical models

An analytic model contains the mathematical relationships and equations that show the specific, quantifiable elements of the model. To further illustrate the purpose and scope of Analytical models, they can be further classified into dynamic or static. Dynamic models describe the varying state of the systems whereas a static model may represent specific values such as mass properties, or the reliability prediction of a component. 

A hybrid between Analytical and Descriptive models is often used to provide the logical relationships of a descriptive model that has been analysed and quantified as part of the system.

Domain Specific Models

To further demonstrate the scope and purpose, illustrating the specific domain will add additional clarity. OWL, Ontologies and SysML Profiles: Knowledge Representation and Modelling (Web Ontology Language (OWL) & Systems Modelling Language (SysML)) (Jenkins 2010) provides additional classifications:

  • properties of the system, such as performance, reliability, mass properties, power, structural, or thermal models;
  • design and technology implementations, such as electrical, mechanical, and software design models;
  • subsystems and products, such as communications, fault management, or power distribution models; and
  • system applications, such as information systems, automotive systems, aerospace systems, or medical device models.

Integrating Models

The terminology, classification, and approach will additionally need to be adapted to the specific model being prepared. In fact, a single model will often involve multiple elements of physical, analytical, descriptive and systems models with a hybrid of logical and descriptive narratives in order to clearly represent the system being modelled. 

Each approach, working together, will represent different facets to more effectively communicate the purpose and scope of the systems.

Additionally, research the modelling standards for modelling languages, transformations and data exchange to enable integration across modelling domains.

Leave a Comment