Fast project delivery is a common goal for all participants (owners, constructors, vendors, etc.) within the construction industry. Prefabrication is attractive because it offers an approach to complete construction projects within a much shorter duration than traditional construction methods (Retik 1994). However, the successful completion of prefabrication projects requires careful planning prior to implementation. The complex nature of prefabrication project scheduling makes traditional scheduling tools (CPM, PERT, Gantt charts, etc.) ineffective due to a common phenomenon termed iteration. Iteration is a cyclic process and results from activity interdependency. Dependency Structure Matrix (DSM) is an alternative approach that has had much success in manufacturing but has seen limited application in construction. It offsets the shortcomings of the traditional scheduling tools in dealing with the phenomenon of iteration. This paper first discusses the applicability of DSM to prefabrication projects. An example model that demonstrates the application of DSM is then developed and used to demonstrate determining an optimal schedule in terms of total feedback distance. To improve the accuracy of project duration estimations, the Reward Markov Chain method is employed to estimate durations of sequential iterations. In addition, an example of using the Reward Markov Chain method is demonstrated. The paper concludes with a discussion of the advantages and disadvantages of the approach.