A mathematical programming approach for generating alternatives in discrete structural optimization

Structural design, like other complex decision problems, involves many tradeoffs among competing criteria. While mathematical programming models are increasingly realistic, there are often relevant issues that cannot be easily captured, if at all, in a formal system. This paper describes an approach to modelling that recognizes these limitations and allows a designer to explore unmodelled issues in a joint human–computer cognitive system. A prototype based on this approach is presented for topological truss optimization, and three modelling techniques are contrasted for their effectiveness in producing ‘different’ alternatives. The results show that alternatives produced using these techniques are good with respect to modelled objectives, and yet are different, and often better, with respect to interesting objectives not present in the model.