This work presents a powder bed-based additive manufacturing technique for the consolidation of Ti-6Al-4V powders onto commercially pure titanium plates. Experiments were conducted to investigate and explain the influence of various process parameters, including laser power (100-200 W), exposure duration (50-175 μs) and point distance (35-70 μm), on the material density and porosity of the printed body; as well as the dimensional accuracy. A process parameter array was created using fractional factorial tests (Taguchi L9) and the significance of the operating factors was analysed using ANOVA. Thermally induced deflections in the Ti plates were influenced by the volumetric energy density, variations in thermal conductivity, scanning strategy and the layup orientation in laser scanning. Furthermore, the geometry of the printed body was shown to alter the thermal stress relief, leading to asymmetric deformation of the Ti plate. Statistical regression models were used to balance the contradictory criteria of material density and dimensional accuracy (Z1 variation). The lowest dimensional variation (Z1: 0.12 mm) was produced using equal weightings (50%) for the factors dominating material density and dimensional accuracy, respectively. In contrast, setting weighting values of 80% for material density and 20% for dimensional accuracy resulted in the highest material density (4.13 g/cm3).