Molecular Dynamics (MD) Simulations are essential in modern science because they allow researchers to study the behavior of molecules at an atomic level. This has a variety of applications, starting with the property prediction of certain materials and
ending with assistance in the development of newer chemicals. In MD Simulations the position of each molecule is shown at every point in time, which arises from calculations according to the classical laws of physics. With a high number of particles, this can
quickly become computationally expensive and infeasible, which is why researchers are constantly developing newer optimizations for these dynamical systems.

In this thesis, we will focus on three-body interactions with a defined cut-off distance, allowing for more accurate results compared to the conventional approach. In addition to reviewing the existing algorithms for three-body interactions, we will introduce several
techniques built on the linked cell structure in a shared memory environment. In our experiments, we will rely on the existing work for two-body interactions and explore various multithreaded approaches for triplets of particles. Afterward, we will analyze
how the additional optimizations have affected the performance of our simulations compared to the naive approach in shared memory environments. Outlining the advantages and disadvantages allows one to choose the most optimal approach for the
given system and directly present its possible use cases. Finally, we will also discuss the potential applications and future directions of optimized three-body interaction techniques.