Multiobjective Fixed Product Flexible Shop Scheduling with Transportation Considerations
Open Access
- Author:
- Trail, Casey D
- Graduate Program:
- Industrial Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- October 02, 2012
- Committee Members:
- Jose Ventura, Dissertation Advisor/Co-Advisor
Jose Ventura, Committee Chair/Co-Chair
Arunachalam Ravindran, Committee Member
Paul Griffin, Committee Member
Shashi Phoha, Committee Member - Keywords:
- Multiobjective optimization
scheduling
integer programming - Abstract:
- This dissertation introduces the problem of multiobjective fixed product flexible shop scheduling with transportation considerations (MFFST). The MFFST problem consists of locations, tasks that need to be performed at these locations, and workers who must travel between locations in order to perform these tasks. The amount of time it takes each worker to perform each task is different. Furthermore, there is a cost associated with performing each task and the workers' costs differ across both tasks and locations. A schedule to the MFFST determines the worker that performs each task at each location, and includes the routes for workers through the locations to which they are assigned. The objective is to identify schedules that provide optimal tradeoffs between the time required to perform the schedule and the cost incurred by performing every task at every location. The MFFST problem generalizes well known operations research problems such as the traveling salesman problem and the vehicle routing problem. It also generalizes flexible shop scheduling, fixed product shop scheduling, and some versions of the MFFST generalize open shops, job shops and flow shops by assuming a precedence structure for tasks at each location. Fixed product scheduling involves scenarios where the processors move while the product stays in place. This aligns with the assumptions of the MFFST as the workers move between locations to perform tasks. This type of scheduling is especially applicable in the assembly of large vehicles, the construction of physical structures, and the performance of operations across geographical areas. Three versions of the MFFST are introduced in this dissertation. These methods incorporate objectives such as minimizing total cost, minimizing total travel and processing time, and minimizing the makespan of the busiest worker. Additionally these versions consider open shop and flow shop precedence structures for tasks at each location. For each of these versions is formulated as a 0-1 MILP, and heuristic methods that are effective and computationally efficient are presented.