Increasingly powerful workstations and PCs, interconnected through various networks, continue to proliferate throughout the world. Most are greatly underutilized and thus represent a significant computational resource, which could be tapped for running applications requiring large amounts of computations. Typical examples of such computation-extensive application are scientific computations for solving complex problems, optimization applications in the A.I. field, as well as large simulations. Distributed computing can improve the performance by magnitudes for those applications which otherwise need to run days, months, or years on a single computer. The objective of the proposed research is to address the load balancing issue in the distributed implementation of individual-based applications. This requires solving the following problems: 1) Decompose of the simulation space into sub-areas that can be distributed over multiple computers by the underlying mapping and load balancing algorithms. 2) Develop policies to choose and maintain an appropriate set of computers for the application, and develop dynamic load balancing schemes to map decomposed application onto machines that load is balanced while communication overhead is minimzed.