In the systems development industry, we have been advocating a model-based approach to development for years. We have also preached the benefits of virtual teams and claimed that it was no longer necessary for project teams to be co-located. In March 2008, a team was formed to define a specification for a Unified Profile for DoDAF and MODAF. From the beginning we decided that we were going to live up to our principles by creating a model of our requirements, and generate the profile and the XMI from the model. Discussions and interchange of information were to be based on the model information and all information was to reside in the model. The military organizations involved were the US DoD, DISA, Mitre, the UK MOD, Canada DND, NATO, and Generic AB for the Swedish Military. UML Tool Vendors were Adaptive, Artisan Software (Co-Chair), EmbeddedPlus, NoMagic (Co-Chair), Sparx Systems, and Visumpoint. Industry partners providing domain expertise were BAE Systems, Lockheed Martin, Raytheon, and Rolls Royce. Advisors were Decisive Analytics, Silver Bullet, Model Futures, and ASMG Ltd. The team was a distributed multi national team with representatives from the US, UK, France, Sweden, Lithuania, Australia, Canada, and Thailand in 11 different time zones.
So what is UPDM and why is it necessary? A plethora of military architectural frameworks, such as DoDAF, MODAF, NAF, DNDAF, MDAF and ADOAF is emerging. Each one adds to, redefines and/or clarifies the concepts, views, viewpoints and concerns contained within Military Architectural Frameworks, with the intention of improving procurement, planning, and implementation of military systems. However, supporting multiple and sometimes divergent frameworks leads to problems for industry, military organizations and tools vendors alike. In this age of globalization, mil-aero companies provide systems across the world to multiple governments. Often they must be specified in the local architecture framework creating extra overheads. Incompatible frameworks cause interoperability problems between governments because models cannot be exchanged. Interchange, even between modelling tools supporting the same framework, is difficult, if not impossible due to the different underlying implementations. Finally, having to support several constantly changing framework formats means that modelling tool vendors have a support nightmare.
The Unified Modelling Language (UML) and Systems Modelling Language (SysML) used as an underlying mechanism for all of these frameworks makes it feasible to work towards a standardized UML/SysML profile for these Military Architectural Frameworks. It is important to stress that UPDM is not a new architecture framework. Instead, it provides a consistent, standardized means to describe DoDAF and MODAF architectures in UML-based tools as well as a standard for interchange.
The rest of the paper will not look at UPDM itself, but will deal with the organization of the teams and sub-teams, how we worked and didn't work together, lessons learned and what we will change in the future. It is worth noting that the UPDM specification passed through all the votes during the September and December 2008 OMG meetings. Consequently, the project has been a success. We will also have a chance to improve our ways of working as UPDM phase 2 will attempt to include DNDAF, NAF 3, Human Factors Views and DoDAF 2.0.
\* הוצג בכנס הנדסת מערכות מרץ 2009 –תיכון מערכתי ליתרון תחרותי
 

לצפייה בקובץ