For or against the integration of a batch engine?

Most integrators in our industry believe that integrating a batch engine results in higher costs in terms of licenses and development.
Also, unless the end user requests a "batch engine" or a "ISA-88 standard", they often prefer to carry out additional specific developments, on the basis of a supervisor, in which they only use the recipe mechanism.
Is it justified ?

With conventional solutions, their position is legitimate.

batch processing

The cost of licenses

Although their functional richness is important, licenses of batch engines on the market are generally expensive and not very progressive. "Essential options" can be added, which raises the cost even higher (additional redundant server license, logging option on a standard database, etc. ).

Heterogeneity of software used

In the end, the control solution proposed is based on two, three (or more) different software programs, originally developed by different companies. These programs use separate terminologies and concepts, and do not have the same ergonomics. Therefore, staff training is cumbersome and even after this training, staff is required to go back and forth constantly between the different concepts.
Unless we carry out many projects of this kind, maintaining operational competence is almost impossible, which imposes problems for maintaining projects.

A low level of integration

Because of their different origin, software programs are not sufficiently integrated. The benefit brought by the functional richness of the different modules does not make up for the complexity of integration, sometimes by a great deal. Many specific developments are then necessary to "glue" the different modules together.