Laser aided additive manufacturing (LAAM) is one of the key metal 3D printing technologies for surface cladding or fabrication of near-net shape parts. The study of LAAM scanning pattern is important to understand their relationship with residual stress and part distortion. This paper proposed a framework to both simulate and evaluate laser scanning paths. Firstly, an efficient 3D thermal history analysis finite element (FE) model was developed to predict temperature field evolution for arbitrary scanning patterns. Subsequently, a thermal field based evaluation method was established to determine the optimal scanning pattern with minimal distortion. The effectiveness of the framework was validated experimentally by depositing a rectangular clad on a cuboid substrate with five different scanning patterns. Experiments showed width-wise Zigzag scanning pattern yielded largest distortion. This method with 5 criteria and two evaluation levels were effective to identify an improved width-wise Zigzag scanning pattern by adjusting the deposition paths sequence, while determining the length-wise scanning as the optimal pattern. This work also demonstrated that the temperature field can be used to make qualitative evaluation of LAAM induced distortion which further reduces computational costs.