Computer simulations have played an important role in understanding tribological processes. They allow controlled numerical "experiments" where the geometry, sliding conditions and interactions between atoms can be varied at will to explore their effect on friction, lubrication, and wear. Unlike laboratory experiments, computer simulations enable scientists to follow and analyze the full dynamics of all atoms. Moreover, theorists have no other general approach to analyze processes like friction and wear. There is no known principle like minimization of free energy that determines the steady state of non-equilibrium systems. Even if there was, simulations would be needed to address the complex systems of interest, just as in many equilibrium problems.Tremendous advances in computing hardware and methodology have dramatically increased the ability of theorists to simulate tribological processes. This has led to an explosion in the number of computational studies over the last decade, and allowed increasingly sophisticated modeling of sliding contacts. Although it is not yet possible to treat all the length scales and time scales that enter the friction coefficient of engineering materials, computer simulations have revealed a great deal of information about the microscopic origins of static and kinetic friction, the behavior of boundary lubricants, and the interplay between molecular geometry and tribological properties. These results provide valuable input to more traditional macroscopic calculations. Given the rapid pace of developments, simulations can be expected to play an expanding role in tribology.In the following chapter we present an overview of the major results from the growing simulation literature. The emphasis is on providing a coherent picture of the field, rather than a historical review. We also outline opportunities for improved simulations, and highlight unanswered questions.We begin by presenting a brief overview of simulation techniques and focus on special features of simulations for tribological processes. For example, it is well known that the results of tribological experiments can be strongly influenced by the mechanical properties of the entire system that produces sliding. In much the same way, the results from simulations depend on how relative motion of the surfaces is imposed, and how heat generated by sliding is removed. The different techniques that are used are described, so that their influence on results can be understood in later sections.The complexities of realistic three-dimensional systems can make it difficult to analyze the molecular mechanisms that underly friction. The third section focuses on dry, wearless friction in less complex systems. The discussion begins with simple one-dimensional models of friction between crystalline surfaces. These models illustrate general results for the origin and trends of static and kinetic friction, such as the importance of metastability and the effect of commensurability. Then two-dimensional studies are described, with an emphasis ...