Cost Optimization of Welded Plate Girders

by Natarajan Annamalai, Sr. Design Engr.; Everest Engrg. Works, Cross Cut Road, Coimbatore, Madras, India,
Albert D. M. Lewis, Assoc. Prof. of Struct. Engrg.; Purdue Univ., Lafayette, IN,
John E. Goldberg, Prof. of Struct. Engrg.; Purdue Univ., Lafayette, IN,

Serial Information: Journal of the Structural Division, 1972, Vol. 98, Issue 10, Pg. 2235-2246

Document Type: Journal Paper


Procedures are developed for the optimum design of welded plate girders by the 1969 Specification of the American Institute of Steel Construction. The problem is a nonlinear optimization problem in which many of the design variables are defined only at discrete value because steel plates are available only in certain increments of width and thickness. Simple supports are assumed, and loading may consist of a combination of a uniformly distributed load and concentrated loads. The objective function is the estimated cost of the girder. This function is not unimodal. Backtrack programming is used in selecting the optimum girder. The vector of design parameter values is built up one component at a time by testing parameter values in modified criterion functions to determine if there is a possibility that a more economical girder is available than the best found previously. For symmetrical girders with symmetrical loadings, there is an option to include the investigation of reduced flange plate thickness at the span ends.

Subject Headings: Plate girders | Steel plates | Concentrated loads | Load combinations | Load distribution | Welding | Steel construction | Thickness

Services: Buy this book/Buy this article


Return to search