The Optimal Division of Business Processes Into Subtasks With Specialization and Coordination

This paper analyzes the factors influencing the optimal partition of a business process into subtasks to minimize the process cycle time. The paper develops a single-class open, queueing network model that incorporates the productivity gains due to the division of labor and the server time lost to coordination. An optimization model to minimize the cycle time subject to a maximum number of servers is presented. We derive an upper bound on the number of subtasks and present an algorithm to optimally solve the problem. We conduct an experimental analysis using the model to understand how the coordination time, process technology, and utilization affect the optimal division of labor. The model confirms that as the coordination time increases, the optimal number of subtasks decreases. Moreover, the results show the technology parameter has little effect when coordination is high, and if the utilization increases, then the optimal number of subtasks increases. The model and results provide a means to determine the optimal number of subtasks for a process.