Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Industrial Engineering

Committee Member

Dr. Mary E. Kurz, Committee Chair

Committee Member

Dr. David Neyens

Committee Member

Dr. Kevin Taaffe

Committee Member

Dr. Joshua Summers


Assembly lines are cost efficient production systems that mass produce identical products. Due to customer demand, manufacturers use mixed model assembly lines to produce customized products that are not identical. To stay efficient, management decisions for the line such as number of workers and assembly task assignment to stations need to be optimized to increase throughput and decrease cost. In each station, the work to be done depends on the exact product configuration, and is not consistent across all products. In this dissertation, a mixed model line balancing integer program (IP) that considers parallel workers, zoning, task assignment, and ergonomic constraints with the objective of minimizing the number of workers is proposed. Upon observing the limitation of the IP, a Constraint Programming (CP) model that is based on CPLEX CP Optimizer is developed to solve larger assembly line balancing problems. Data from an automotive OEM are used to assess the performance of both the MIP and CP models. Using the OEM data, we show that the CP model outperforms the IP model for bigger problems. A sensitivity analysis is done to assess the cost of enforcing some of the constraint on the computation complexity and the amount of violations to these constraints once they are disabled. Results show that some of the constraints are helpful in reducing the computation time. Specifically, the assignment constraints in which decision variables are fixed or bounded result in a smaller search space. Finally, since the line balance for mixed model is based on task duration averages, we propose a mixed model sequencing model that minimize the number of overload situation that might occur due to variability in tasks times by providing an optimal production sequence. We consider the skip-policy to manage overload situations and allow interactions between stations via workers swimming. An IP model formulation is proposed and a GRASP solution heuristic is developed to solve the problem. Data from the literature are used to assess the performance of the developed heuristic and to show the benefit of swimming in reducing work overload situations.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.