The criteria against which the Architectural Proof-of-Concept is to be evaluated are drawn from the architecturally significant requirements, which were the drivers in its construction.

In this step, the Architectural Proof-of-Concept is tested against the evaluation criteria: the way in which this is done will depend on the form of the proof-of-concept. For example, in the case of an executable prototype, this may be through demonstration; in the case of a conceptual model, through inspection and reasoning, or, for a simulation, requiring the set-up and running of the simulation model with input data derived from the evaluation criteria, then the collection and analysis of output data from the model.

The results from the evaluation are assessed to determine not only if the architecturally significant requirements can be satisfied, but also as a check on the validity of those requirements. At this time in the development, requirements are still mutable, and not necessarily well-understood by the stakeholders; for example, perhaps the opportunity exists to relax requirements that were shown to be high-risk by the evaluation of the Architectural Proof-of-Concept. All these avenues should be thoroughly explored in assessing the results - this contrasts with the situation later in elaboration and construction, when there will be much greater reluctance to change or reinterpret requirements. After the assessment, with a better understanding of scope and feasibility by all stakeholders, change proposals to the Business Case, Vision and Risk List are prepared, if necessary.

In addition to the above concerns, answering the following questions can also be useful:

• How does the prototype stack-up against its original goals and expected schedule ?
• How does the design of the prototype compare against the Architecture baseline?
• What is our new understanding of the project risks (whether or not they were directly addressed by the prototype)?