The application of simulation in building design and guidance on modelling approaches

Consider the following scenarios: The WWW page describes the tool as "a general first-principles solver for multiple domain assessments" and the practitioner says "the project is an aluminium smelting plant so what's the problem with 3kW/mˆ2 of casual gains?" Tool developer raises eyebrows - simulation at a newly defined edge. Gripe from simulationist to tool developer - "we just finished a project with 100,000mˆ2 floor area and found it difficult to QA the project files". Tool developer imagines scenes from Dante's Inferno - simulation as an endurance contest. Conversation between tool developer and simulationist - "how do I model stair treads? - where are they? - in the fire escape - why do you think this is important? - they are included in the building section". Simulation drowning in needless complexity. Planning discussion "the client wants to know how many minutes they have before environmental conditions deteriorate if there is a catastrophic failure of the HVAC on a sunny summer day - if it goes above 26C a collection of rare artifacts will suffer irreversible damage. How shall we approach this?". Gasp - truly simulation at the edge. This paper reviews some of the singularly amazing, useful, and sometimes misdirected work that practitioners attempt with simulation and the technical and software support which ESRU (the Energy Systems Research Unit) and SESG (the Scottish Energy Systems Group) in Glasgow have recently provided. The paper will deal with issues such as: What was done to support substantial simulation questions and who is well placed to attempt such work? How does one gain confidence in projects which are exploring untested uses of a tool? What duty is it of the vendor to deprive the practitioner of their chaos and complexity? Sometimes it is necessary to build models which are at the compositional limits of a tool. But does the practitioner realize the incremental cost of such complexity? How do those entering the profession develop the skills to deal with issues of model abstraction in realistic design projects? What about when the practitioner has a different idea of what the tool is capable of from that of the tool developer?