A model approximates reality to make a complex situation analysable and may be physical, software or mathematical. Modelling involves describing the problem and purpose, building a conceptual model, building a representation, developing analysis tools, processing data, validating or calibrating against known situations and drawing conclusions. Steps involve approximations, assumptions and judgement that are validated, ideally by people independent of the developers, and critical assumptions are reviewed. Reliable results require validating that the model fits the situation and is used within its limits, that theory, parameters and mathematics are understood, that input data is reliable, that it runs without errors and is stable under small input changes, using sensitivity analysis, stress scenarios, comparison with other data, independent runs and checks against actual performance. Documentation is kept to allow validation.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.