A Heuristic-Based Framework for Cost-Effective Product Design Enabled by Manufacturing Automation: Application to Large-Scale Sheet Metal Structures

Abstract

Sheet metal is prominent as a raw material for fabrication due to its flexible nature. Through cutting, bending, and joining, it can take a plethora of shapes, explaining its vast adoption in the construction, automotive, and aerospace industries. Furthermore, with automation, the labor and human error associated with its manufacturing can be mitigated. Nonetheless, the versatility of sheet metal can fade under the non-trivial dimensional and thickness constraints of some automated processes, particularly bending. This research, conducted in the context of a large-scale sheet metal manufacturer offering high customization, aims to maximize sheet metal’s automation capabilities while retaining its flexibility. To achieve this, two approaches are used: 1) the adoption of rollformed steel profiles with automated tube laser cutting as an additional manufacturing value stream, and 2) the development of a design automation tool that, upon receiving the dimensions and structural load conditions of a rectangular prism (called sub-module), generates a low-cost, automation compliant design. Findings show that optimal modules generally use medium to low-gauge channels as connected structural members, and thin-gauge sheet metal panels as slabs and shear walls, minimizing material use: the main cost component. Generated designs show cost reductions of up to 32% when compared to legacy counterparts. For the most produced product, this translates to yearly cost savings that range from $1.7 to $5.2 million.