In this research, an interactive, computational and experimental programme is presented that permits the rapid design of an optimum array of air-jet vortex generators (AJVGs) for use on an unswept, three-element high lift geometry in low speed flow. The interactive regime relies on the transfer of information between computational and experimental investigations, monitoring the flow field characteristics around the three-element high lift system, with the AJVGs active and quiescent. Careful definition of a 'local' flat plate AJVG computational model enables an assessment of the beneficial trends associated with the AJVG geometry and array configuration, prior to installation on the multi-component aerofoil system. Two numerical approaches are employed to predict the flow field characteristics. The first combines a 2-D streamline based Euler solver coupled with an integral lag-entrainment boundary layer method (MSES of MIT). The second employs a finite volume blockstructured, full Navier-Stokes (NS) flow solver, utilising the k-ε turbulence model with wall functions (CFX4 of AEA Technology). Computational studies of the flow field characteristics with the AJVGs active on the 'local' flat plate model and on the three-element high lift system utilise the NS solver only. Wind tunnel tests were conducted on the three-element high lift system for a range of angles of attack (0°< α < 36°) and jet blowing momentum coefficients (0.0 < Cμ < 0.12) in City University's T2 low speed wind tunnel (Rec=1.37xl0⁶). The high lift system was configured with the slat and flap high lift devices in representative take-off settings with an array of 13 equi-spaced co-rotating AJVGs across the span of the wind tunnel model, located at x/c= 0.14c. The wind tunnel experiments involved measuring the chordwise surface static pressure and skin friction distributions and the wake profile downstream of the trailing edge. The trends observed from the 2-D high lift system computations agreed with the experiments up to α≈20°, subsequent to which the CFX4 predictions deviated from the MSES and experimental findings. These differences result from neglecting the effects of boundary-layer transition; and the reliance on the law of the wall in the NS computations that is invalidated where flow separation is substantial. Improvements in the NS predictions can only be expected by making adequate provision for boundary-layer transition and by better modelling of the turbulence in the confluent boundary layers and near-wall flows above the wing component surfaces. Computational models with the AJVGs active are able to represent the experimentally observed trends in flows where the adverse pressure gradient is weak. Using the 'local' AJVG model to identify beneficial trends in the flow field characteristics (filling out of the streamwise velocity profiles and enhanced skin friction), near-optimum air-jet spacings and flow features were determined. In stronger pressure gradient flows, however, more robust definition of the numerical boundary conditions either side of the air-jets is required, to facilitate adequate representation of the flow field downstream of the AJVG arrays. Experimental results for the high lift system incorporating the AJVGs show it is possible to delay the onset of flow separation by up to 7° angle of attack, increase the maximum normal force generated by the high lift system by 25% and significantly delay drag rise. The greatest normal force enhancement ΔCNmax=0.6, relative to the uncontrolled flow case, was determined with Cμ=0.057 but useful flow control was achieved at values of Cμ as low as 0.014.