Entropy-based optimum sizing of trusses

A new method for optimally sizing the bars of a structural truss is described. Though many different methods for solving this problem have been proposed in the past, the present method is radically different in that it sets the problem in a probabilistic context and uses the Shannon entropy function to infer improved solutions based upon available information. The resulting method is extremely simple to operate and is efficient, suggesting that there may be other potential uses for informational entropy in applications not usually concerned with statistical inference.