Efficient Material Utilization in Truss Structures Using Deterministic and Metaheuristic approach

Abstract

When the geometry and connectivity of a truss are fixed, efficient material utilization under strength, serviceability, and stability constraints is addressed as a size optimization problem. This study presents a comparative study of the material minimization of a truss structure by using three optimization algorithms: Sequential Quadratic Programming (SQP), Genetic Algorithm, and Particle Swarm Optimization (PSO). For comparative analysis to keep same framework, these methods are employed to optimize a benchmark 10 Bar truss problem. The minimization of material is achieved by determining optimized area of cross-section of truss members. For comprehensive comparative analysis between optimization methods, statistical key performance indices are evaluated by focusing on optimized area and overall weight of truss, integrating both serviceability (Stress and displacement constraint) and practical sustainability (frequency constraint) criteria. While achieving material minimization, convergence behaviour and computational efficiency are analysed through weight iteration histories and execution time. The results indicate that SQP yields the minimum structural mass of 2486.88 kg and also the lowest Total Structural Area (TSA) of 0.0825 m². The ranking score shows that both SQP and PSO have better performance in convergence and accuracy. In contrast, the GA produces higher structural mass, greater displacement in member, and larger error values. SWOT analysis of all these 3 approaches helps for a better understanding of working mechanism, applicability, accuracy. Which guides for selection of optimization method. The SQP method showed the best accuracy overall. However, study reflects PSO is a good and reliable metaheuristic alternative for optimization of truss.