Feedback

It will be interesting to see how well your choice of topic turns out as an exemplar for an EM approach. Dependency is clearly evident in some aspects of the drivetrain simulation, but it may be technically quite hard to model this as elegantly and attractively as you'd like using current tools. Possible issues are dealing with the changing of dependencies that may have a major impact on structure (gear changing is an archetypal example of an action that changes a dependency relation and might have significant geometric impact in this context) and getting a satisfactory visualisation. Aiming your article at newcomers is probably a shrewd move, but you may need to be prepared to be quite critical of EM tools etc in some aspects! Perhaps you should substitute 'strengths and weaknesses' for 'strengths' in your third paragraph, even though, if our principles and tools were all we'd like them to be, agent-oriented modelling would subsume the best features of object-oriented modelling.

Good to be more specific about the references you might use. It's not entirely clear from your discussion whether a dynamic model of the degree of complexity of the VCCS is envisaged here. For instance, it is clear that speed is one relevant output, but presumably the force that must be applied at the pedals to achieve a specific speed in a given gear and context would also be of interest. Some of this modelling might take you outside the scope of a first case study. Describing visualisations of sprockets etc is also fairly challenging in donald, and it might be that you could consider using the AOP to introduce some kind of auxiliary notation for specifying such things. (This could suit the 'illustrating EM in a case study' theme quite well.)

Your abstract and description of the model-building is still somewhat in note form, and it will of course be important to refine your conception and improve the narrative and continuity in due course.