Data-Driven Multi-Response Optimisation of WAAM-Fabricated SS316L Using Evolutionary and Bio-Inspired Algorithms

The demand for large-scale metal components has emerged, with structural integrity, material waste optimisation, and superior operational flexibility driving the adoption of wire arc additive manufacturing (WAAM) in extreme applications. Stainless steel, specifically SS316L, is a widely used alloy and can be deposited efficiently using WAAM, enabling near-net-shape components. However, the properties of WAAM-fabricated SS316L components are primarily controlled by deposition quality and heat input. The bead geometry and surface morphology also play a crucial role that affects structural precision, interfacial bonding, and secondary processing needs that further govern the in-service performance. Therefore, this research aims to develop a multi-objective process optimisation architecture to identify optimum process parameters of WAAM-fabricated SS316 using 27 experimental data points. The effect of current (I), travel speed (TS) and interlayer time (ILT) on three responses, such as bead width, bead height and surface roughness, was examined systematically to identify consistent deposition morphology with enhanced surface properties. Two metaheuristic algorithm-based optimisation techniques, such as the Genetic algorithm (GA) and African vulture optimisation algorithm (AVOA), are utilised to determine the optimum process settings. The optimal I (100 A), TS (7 mm/s), and ILT (78 s) are identified with minimal heat input to enhance geometric and surface morphology simultaneously. The AVOA ensures higher efficiency with an improved overall mean square error (MSE) of 5.11% compared to GA (MSE = 5.63%). The outcomes demonstrated the potential of a metaheuristic-based optimisation approach for WAAM-fabricated SS316L alloy.