How publishing works in this field
The field publishes across journals and, in some subfields, archival conferences. IEEE and Elsevier titles dominate much of it, each with its own template, reference style and figure requirements, and editorial offices return non-conforming submissions before review. Using the current template from the target journal removes an entire class of delay.
Validation is the standard the field applies. Simulation alone is rarely sufficient where measurement is possible; where it is not, the model must be validated against published experimental data or against an analytical limit case, and the paper has to say which and why.
What slows papers down here
Weak or missing comparison is the main rejection cause. Benchmarking against an old method, or against an implementation of a competing method that the authors tuned poorly, is visible to referees working in the area and is treated as a serious flaw.
Reproducibility of the setup is the other. Component tolerances, sampling rates, controller gains, mesh density and solver settings all change results, and a methods section that omits them leaves reviewers unable to judge whether the improvement is real.
Worth reading next
Questions we are asked in this field
In some venues, yes, but it has to be justified: either experimental validation is genuinely impossible, or the model is validated against published measurements or a known analytical case. Reviewers reject simulation presented as if it were measurement.
Pick the strongest current methods in the application area, implement them fairly, and state the settings used. Comparing against a weak baseline is the fastest way to lose a referee who knows the field, and it undermines a result that may have been genuinely good.
It varies sharply by subfield. In parts of computing and communications, selective conferences are the main archival venue; in most of mechanical, civil and chemical engineering, journals carry more weight. Check what your own subfield cites.