Past literature shows that workflows will be performed with greater efficiency and/or effectiveness if workflow teams have higher group cohesiveness. The major contribution of this work in progress is the creation and implementation of a formal generalized methodology that incorporates ideas from two diverse fields: social network theory and workflow modeling, and allows optimization of work groups along group cohesiveness. In order to implement this model we present newly created algorithms to structure and represent the problem of workflow load representation, possible team sets and social network metric optimization so that standard integer programming solvers can attempt to solve it.