Reducing the cost of concrete structures using RBDO

Abstract
The construction industry is based on reinforced concrete. From foundations to walls and ceilings, reinforced concrete is the most widely used material in the construction sector. Despite its excellent properties, concrete has a big negative impact on people and the environment. 

The aim of this thesis is to determine the reduction potential of concrete consumption of reinforced concrete structures with reliability-based design optimization (RBDO), a probabilistic approach. For this purpose, two simplified reinforced concrete structures and a punching problem were designed each semi-probabilistically and probabilistically and then compared. It is shown that reinforced concrete structures, especially slab systems, have considerable reduction potential due to the conservative semi-probabilistic design. Depending on the system, concrete type and reinforcement content, reductions of up to more than 25% of concrete volume are possible.

Aware of the reduction potential of slabs, the goal is to generalize reliability-based design optimization of a generic slab structure to account for different configurations without having to evaluate each configuration individually. In this context, graphical tools based on surrogate models were created, which are expected to make probabilistic design quick and easy to apply.

This work has highlighted the potential gains that can be made by a thorough probabilistic design. However, there are still some hurdles hindering the application of such techniques in the day-to-day operations of structural engineers. These mainly include considerable uncertainty regarding live loads, the lack of standards and the lack of universal agreement on factors, safety levels and the distributions of parameters.