Skip to main content

Conceptual design optimization of rectilinear building frames: a knapsack problem approach

Journal Article


Abstract


  • This article presents an automated technique for preliminary layout (conceptual design) optimization of rectilinear, orthogonal building frames in which the shape of the building plan, the number of bays and the size of unsupported spans are variables. It adopts the knapsack problem as the applied combinatorial optimization problem, and describes how the conceptual design optimization problem can be generally modelled as the unbounded multi-constraint multiple knapsack problem. It discusses some special cases, which can be modelled more efficiently as the single knapsack problem, the multiple-choice knapsack problem or the multiple knapsack problem. A knapsack contains sub-rectangles that define the floor plan and the location of columns. Particular conditions or preferences for the conceptual design can be incorporated as constraints on the knapsacks and/or sub-rectangles. A bi-objective knapsack problem is defined with the aim of obtaining a conceptual design having minimum cost and maximum plan regularity (minimum structural eccentricity). A multi-objective ant colony algorithm is formulated to solve the combinatorial optimization problem. A numerical example is included to demonstrate the application of the present method and the robustness of the algorithm.

Publication Date


  • 2015

Citation


  • Sharafi, P., Teh, L. H. & Hadi, M. N. S. (2015). Conceptual design optimization of rectilinear building frames: a knapsack problem approach. Engineering Optimization, 47 (10), 1303-1323.

Scopus Eid


  • 2-s2.0-84938747052

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers/3761

Has Global Citation Frequency


Number Of Pages


  • 20

Start Page


  • 1303

End Page


  • 1323

Volume


  • 47

Issue


  • 10

Place Of Publication


  • United Kingdom

Abstract


  • This article presents an automated technique for preliminary layout (conceptual design) optimization of rectilinear, orthogonal building frames in which the shape of the building plan, the number of bays and the size of unsupported spans are variables. It adopts the knapsack problem as the applied combinatorial optimization problem, and describes how the conceptual design optimization problem can be generally modelled as the unbounded multi-constraint multiple knapsack problem. It discusses some special cases, which can be modelled more efficiently as the single knapsack problem, the multiple-choice knapsack problem or the multiple knapsack problem. A knapsack contains sub-rectangles that define the floor plan and the location of columns. Particular conditions or preferences for the conceptual design can be incorporated as constraints on the knapsacks and/or sub-rectangles. A bi-objective knapsack problem is defined with the aim of obtaining a conceptual design having minimum cost and maximum plan regularity (minimum structural eccentricity). A multi-objective ant colony algorithm is formulated to solve the combinatorial optimization problem. A numerical example is included to demonstrate the application of the present method and the robustness of the algorithm.

Publication Date


  • 2015

Citation


  • Sharafi, P., Teh, L. H. & Hadi, M. N. S. (2015). Conceptual design optimization of rectilinear building frames: a knapsack problem approach. Engineering Optimization, 47 (10), 1303-1323.

Scopus Eid


  • 2-s2.0-84938747052

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers/3761

Has Global Citation Frequency


Number Of Pages


  • 20

Start Page


  • 1303

End Page


  • 1323

Volume


  • 47

Issue


  • 10

Place Of Publication


  • United Kingdom