An area of metal additive manufacturing (AM) that has gained significant traction in the last decade is the control of scan strategies to manipulate thermal gradients and develop desired microstructures. This approach offers the potential for engineering crystallographic and morphological features at specific locations to enhance the quality of the manufactured part. Multi-spot scan strategies used in powder bed fusion (PBF) have emerged as promising techniques to achieve desirable microstructural features. This is due to the flexibility offered by adjusting spot locations to manipulate thermal gradients between melt pools. However, the influence of the newly developed multi-spot scan strategy on microstructure evolution is highly dependent on the geometry of the build. This requires a trial-and-error experimental approach to develop new multi-spot scan strategies. Alternatively, computational tools offer the possibility of investigating and optimising multi-spot scan strategies to ensure targeted experimental investigations. In this work, a cellular automata (CA) tool is used to simulate the microstructure evolution of Inconel 738 during AM. The tool is firstly validated against experimental data and then used to investigate the influence of different multi-spot scan strategies. Through this investigation, a new algorithm is proposed to control the level of randomness applied to a predefined scan strategy, providing more control over the evolving microstructure. The findings highlight the importance of performing computational investigations to engineer the optimal multi-spot scan strategies to achieve the most desirable microstructure and limit the possible occurrence of lack of fusion defects.