Design optimization of 3D printed concrete elements considering life cycle assessment and life cycle costing

In answer to the increasing demand for sustainable construction, sustainable construction is gaining momentum. However, the optimization methods cannot work effectively for dealing with uncertainty, non-linearity, and conflicting objectives during early stages. Thus, this paper presents an integrated model combining Gaussian Process Regression (GPR), the Firefly Algorithm, and NSGA-III (Non-dominated Sorting Genetic Algorithm III) to optimize 3D-printed construction elements. GPR provides robust predictions that are essential for Life Cycle Assessment (LCA) and Life Cycle Costing (LCC) assessments because it contains uncertainty estimates. It is suited to the search of efficiently finding the global optimum in a nonlinear design space; however, NSGA-III is used in multi-objective optimization by taking care of issues associated with environmental impact, cost, and structural integrity. This work applied this model to 3D-printed concrete beams and showed that it resulted in a reduction in global warming potential by 30–40% and life cycle costs by up to 20%. The results varied from a single notion to approaching some 21,600 optimal solutions with all possible diversities and providing for a wide array of Pareto-optimal designs for sustainable construction. Thus, a new benchmark is established to optimize simultaneously environmental and economic performance of construction elements in 3D printing process.