Date of Award

Fall 2016

Document Type


Degree Name

Doctor of Philosophy (PhD)


Engineering Management & Systems Engineering

Committee Director

Holly A. H. Handley

Committee Member

Resit Unal

Committee Member

Ghaith Rabadi

Committee Member

Roland Mielke


The outcome of a work process depends heavily on which tasks assigned to which employees. However, sometimes-optimized assignments based on employees’ qualifications may result in an uneven and ineffective workload distribution among them. Likewise, an even workload distribution without considering the employee's qualifications may cause unproductive employee-task matching that results in low performance of employees. This trade-off is even more noticeable for work processes during critical time junctions, such as in military command centers and emergency rooms that require being fast and effective without making errors.

This study proposes that optimizing task-employee assignments according to their capabilities while also keeping them under a workload threshold, results in better performance for work processes, especially during critical time junctions. The goal is to select the employee-task assignments in order to minimize the average duration of a work process while keeping the employees under a workload threshold to prevent errors caused by overload. Due to uncertainties inherent in the problem related with the inter-arrival time of work orders, task durations and employees' instantaneous workload, a utilized simulation-optimization approach solves this problem. More specifically, a discrete event human performance simulation model evaluates the objective function of the problem coupled with a genetic algorithm based meta-heuristic optimization approach to search the solution space.

This approach proved to be useful in determining the right task-agent assignments by taking into consideration the employees' qualifications and mental workload in order to minimize the average duration of a work process. Use of a sample work process shows the effectiveness of the developed simulation-optimization approach. Numerical tests indicate that the proposed approach finds better solutions than common practices and other simulation-optimization methods. Accordingly, by using this method, organizations can increase performance, manage excess-level workloads, and generate higher satisfactory environments for employees, without modifying the structure of the process itself.