A data transformation is an object that allows you to copy data from a source container to a target container, and also to calculate the value of an expression and store it in a variable.
For example, a manufacturer may be asked the price of one of its products, which he calculates depending on one or more parameters, such as quantity, delivery location, and so on. These parameters are input data on which the manufacturer will perform a transformation, and then store the result as target data.
A data transformation can be created in a choreography diagram with any orchestration languages.
It is used in conjunction with:
Assign activities – to design a sequence of atomic assign tasks (see Process Properties)
Correlation keys – to perform a mapping between a message, and a variable, which identifies a process instance (for example, a customer ID) (see Correlation Keys (BPM))
A data transformation can have one or more source containers (Input Variable), but has always only one target container (Assigned Variable). In addition, the value to be copied from the source to the target must be type-compatible.