The paper analyses methods of formalising cognitive graphics and visual models using promising data storage formats. We describe the primary visual design techniques and note that they appear to be rather disconnected. We show that ensuring the coupling of data and knowledge in visual models featuring various levels of detail is the main problem in integrated usage of visual modelling tools. We analyse approaches to solving the semantic discontinuity problem, that is, provided we meet the condition under which the properties of objects, systems and processes under consideration are only input once, it is necessary to ensure that data from models corresponding to different levels of abstraction (expertise) is interconnected. One should assume that the main drawback of existing approaches to visualising complex systems is that these approaches are fragmented and isolated, which means that they will only be effective locally. The paper proposes several approaches to formalising visual models employing XML schemas, which ensures that development processes concerning visual models of various levels of abstraction are synchronised and interconnected. We use a BPMN (Business, Process, Model and Notation) visual model as an example that shows the principles of representing visual model elements by means of XML schemas. The paper provides a detailed analysis of the principles behind layer interaction in the BPMN model through flexible XML description. We show that the BPMN data structure boasts its own XML schema containing all the parameters of a class or an element. The paper presents several examples and a technique of applying an XML schema to a BPMN model, including a further generalisation aimed at formally representing the process models of complex systems

No abstract

No abstract

No abstract

No abstract