Employee training and assignment for team-based manufacturing systems
AdvisorAskin, Ronald G.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractCellular manufacturing has been extensively adopted as a measure to reduce cycle time, increase productivity, and improve product quality. The past research in cellular manufacturing has focused on the methodology for identification of machine groups, part families, and determination of processing routes while the relocation of existent workers into cells and their training for cellular manufacturing environment have been ignored. However, several industrial surveys show that human and administrative issues are a major unsolved problem in implementing cells. Human issues in the cell formation process have received little attention in the literature. This dissertation develops methods for guiding the assignment of workers to cells and determination of training plans and task assignments for workers. An integer programming model is first proposed to determine the assignment of workers to cells and the aggregate training needs for each cell with consideration of meeting the technical and administrative skill requirements in each cell. In addition to the technical and administrative skills, team synergy level predicted on the basis of the combination of individual personality-related traits, and individual job fitness are then included in the consideration for building high performance teams. A mixed integer programming model is formulated with objective to create effective manufacturing teams meeting cell requirements with low training cost, high team synergy level, and compatibility between workers and tasks. Several solution methods including greedy heuristics, beam search, filtered beam search, and simulated annealing techniques are developed for solving the mathematical models. They are tested and compared to a standard optimization software for a set of test problems. Results indicate that problem size, initial mix of skills, and the skill requirements of cells in the data set impact the difficulty of obtaining good solutions. Nevertheless, it appears that heuristics such as beam search are capable of obtaining good solutions with reasonable computational effort. Directions for future work are discussed at the conclusion of this dissertation.
Degree ProgramGraduate College
Systems and Industrial Engineering