Be Advised: Classified, FOUO, and PII Content is Not Permitted

  • Replies 1 reply
  • Subscribers 462 subscribers
  • Views 7484 views
  • Users 0 members are here
Related

Primer on Federated Authoritative Models

Jim Ciarcia

We just got Distro A approval for releasing this Primer from our Naval Digital Engineering Body of Knowledge wiki.  The Department of Navy Systems Commands recognize that monolithic, stand alone models are not the way of the future for the highest return on investment in Digital / Model Based Engineering.  This document represents the latest "written" "best practice" for doing federation of MBSE Architecture Descriptions.  It is not the latest best practice, and people would even argue if it represents best practice at all.  So we are asking the community weigh in on the concepts here.

This Primer is exactly what the term 'primer' suggests, a preparatory work that sets the stage for a more detailed refinement of the topic elsewhere.  Unfortunately we don't have that detail sufficiently documented to share it now. What we have documented would still leave large holes in the full understanding of how we see best practice actually applying the federation concepts in this Primer to actual MBSE technology, like SysML.  So this document is meant to stimulate conversation, not to train practical execution of the concepts.  Architects and modelers will still need to bring their expertise of MBSE to bridge that gap for now.

At 12 pages, it is not a long read.  Looking forward to seeing what ideas this primer stimulates.

Naval Primer On Federated Authoritative Models

The below macro view of the document is a little wonky to use on some browsers, the link above should take you to a much larger view of the file.

The link below will let you download the file if you log into your APAN account.  If you don't have an APAN account you can easily create one and request to join the NAVAIR MBSE CoP.  Apologize for having to have people jump through hoops like this, but known alternatives are worse.

Download Link

DOCX

  • I really appreciate any discussion about federating models because I think it is potentially an important technical requirement for a program's MBSE work. It should be thoughtfully planned and will have to take into account the program's classification level, the size of the workforce that will be devoted to the modeling, the scope of the modeling project, and how the program is partitioned. Like you said, this primer has holes in terms of how to apply the principles and the partitioning strategy. Some of the ideas could use more explicit explanation with examples. For example, it wasn't clear to me how to set up the reuse patterns described for the black box and white box models. Also, in the second paragraph, it says that the paper will recommend naming conventions as well as specific metaclasses and stereotypes for various namespaces. I am not sure I found those recommendations in the paper. I am one who would say that it is less a description of a best practice and more an academic intro to some principles and strategies related to federating.