Design Optimization of Reinforced Concrete Frames using Artificial Bee Colony Algorithm

Abstract

The objective of this study is to develop an optimization model that is capable of obtaining the optimum design for reinforced concrete frames in terms of cross section dimensions and reinforcement details. The optimization is carried out using Artificial Bee Colony (ABC) Algorithm, while still satisfying the strength and serviceability constraints of the American Concrete Institute Building Code Requirements for Structural Concrete and Commentary (ACI318M-08). The Artificial Bee Colony (ABC) Algorithm was recently developed by Karaboga(2005) based on the foraging behavior of a honey bee swarm. The ABC algorithm has proved itself as a reliable and robust optimization algorithm in various optimization problems ranging from numerical functions to the optimization of steel trusses. A broader reinforcement detailing scheme was utilized in this study when compared with the previous studies conducted on similar topics: Cut off bars were utilized in beams to reduce overall cost. Additional considerations were taken into account such as joint detailing, shear design as well as various column reinforcement arrangements. Three case studies were considered. The first case was a frame of one bay and one story and had a design space of 7.46 x 1013 possible frame designs and was used as a test frame to obtain the best combination of the ABC algorithm control parameters. Consequently, two frames were studied: a three bay four story frame with a design space of 4.13 x 1036 possible frame designs, and a three bay eight story frame with a design space of 2.98 x 1052 possible frame designs. The results for the three bay four story frame and three bay eight story were compared with a study previously conducted by other researchers who used two different optimization algorithms, namely The heuristic big bang-big crunch (HBB-BC) algorithm, which is based on big bang-big crunch (BB-BC) and a harmony search (HS) scheme to deal with the variable constraint, and The (HPSACO) algorithm, which is a combination of particle swarm with passive congregation (PSOPC), ant colony optimization (ACO), and harmony search scheme (HS) algorithms. The results prove that the ABC algorithm as well as the design variables used in the ABC yielded better results than the previous study: For the three bay four story frame, cost savings of (5.5%) were achieved whereas for the three bay eight story frame, cost savings of (3.5%) were achieved.