Rochester Institute of Technology
Rolling Horizon Workforce Schedule to Assign Workers to Tasks in a Learning/Forgetting Environment
Abstract
dc:description.abstractThe variability in demand across the planning horizon and the presence of heterogeneous workforces where workers learn and forget at different rates make the process of building and managing a workforce challenging. When integrating learning and forgetting functions of workers into workforce scheduling, the previous experience of a worker on a task can have significant impact on productivity. While making assignments over an infinite planning horizon is ideal, the learning/forgetting function significantly increases problem complexity and solution difficulty as the length of planning horizon increases. In this thesis, a multi-period rolling horizon worker-task assignment framework is developed to overcome computational challenges associated with longer planning horizons. The non-linear learning/forgetting function is converted into an equivalent linear form (using an existing technique) to further reduce problem complexity. We design experiments to analyze the optimal planning horizon and the factors that affect it, questions that remain unanswered in literature. After testing the model under different scenarios (varying staffing level, variation in demand, learning rate, forgetting rate and workforce heterogeneity), we conclude variation in demand and staffing level to be the most significant factors in determining the optimal planning horizon. We also see a significant improvement in performance when comparing our proposed multi-period framework against a myopic model, especially in scenarios with higher workforce heterogeneity, higher variation in demand, and faster forgetting rate.
Degree
thesis:*- Name thesis:degree_name
- Industrial and Systems Engineering (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Industrial and Systems Engineering (KGCOE)
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Potluri, Bhanu Krishna
- Contributors dc:contributor
-
- Scott E. Grasman
Subjects
dc:subject × 4Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.rit.edu/theses/9187