Feedforward refers to the transition from briefing to designing in ways that enrich the design with information from the brief (one assumes that a better-informed designer can achieve higher performance than an uninformed one). The first possibility for feedforward that is discussed is the use of the requirements graph in designing, including the development of schematic designs in Visio. After that, another form of feedforward is examined: the connection of the brief database to design representations. This is considered with respect to the various facilities on offer in CAD (AutoCAD) and BIM software (Revit and AutoCAD Architecture).
TopGraph And Design
Using graphs to represent a brief is a clear first step in its translation into a spatial form and hence a useful bridge to designing. It is therefore not surprising that similar representations have been used as a first step in attempts to automate design processes, as mentioned in Chapter 4. However, there are serious limitations in such attempts, primarily due to the highly abstract character of a graph, which eliminates most of the geometric aspects we need in a design (see section From requirements graph to floor plan in Chapter 4). So if you use a requirements graph as a starting point for designing, keep in mind that it should be primarily used in comparisons to spatial design representations – do not try to make a design around a graph.
Importing a graph you have made in Visio in CAD or BIM software is quite easy if you export it first as a DWG file. An increasing number of programs can also insert a Visio file (e.g., as an OLE object, but in most cases there is little you can do with such an object except view it). If you import a DWG version of a Visio graph in, for example, AutoCAD, vertex symbols are translated into polylines (even if they were originally circles). Straight edges become lines and curvilinear ones splines; in short, all turn into objects you can process in AutoCAD. Visio layers are translated into AutoCAD layers, giving you easy means of clustering imported graph components.
In Revit, things are similar with two notable exceptions. The first is that an imported DWG file becomes a single symbol. In order to be able to work with the graph, you have to explode it first (both full and partial explode have the same effect). After exploding you encounter the second difference: Some vertex symbols are exploded in several graphic objects. In particular, the circular vertex symbols we have used in Chapter 4 become two half-circle arcs (interestingly, in AutoCAD they become a polyline consisting of two such arcs – a common problem with exporting circles from in vector graphics, even in the age of interoperability).