Core Concept behind recording production record data via data model

Using the EAV database model for data recording, allows users to build their data models referencing the ISA-88 standard. Litmus Production Record Database is not a batch process control system, but recognizes the validity of the ISA-88 Hierarchy. Which, also in combination with for example the ISA-95 hierarchy, can still be found at the core of many data models and digital twin models.



*Reference: https://www.researchgate.net/figure/ISA-S95-and-ISA-S88-Physical-Hierarchy-Models-1_fig3_285583863

Based on these standards, data for production events are stored based on a named data model. This name or entity in reference to the EAV model is called a Node Name and is the Level 0 of the data models hierarchy, which can have optionally 5 additional Levels.

The Node Name and Levels could be understood as folders and subfolders. Or as Elements and sub elements in an XML structure or as key and sub-key in a JSON structure.

Data are recorded as Item under each Hierarchy Level and could be understood as a file in a folder or sub folder. OR equally they can be understood as the elements content in XML or the value of a key-value pair in JSON.

Users have total freedom to define how many levels the hierarchy is deep and what items are recorded at which level. This does also include if a sub level has other sub levels and os on.

Example for a data model with no sub levels

Example for a data model with one Sub Level:

Example with 2 Sub Levels:

Example for 5 Sub Levels:

While having a lot of freedom to define the data model, there is also some responsibility which comes with this freedom.

As there does exist a high risk of creating a data model which does not represent the actual situation of the event which is to be recorded.

Therefore, the creation of a data model should always be done after clearly identifying the actual process.

The picture below shows how a data model is idealy a translation of the actual production process.