Developing a Scheduling System for a Shared Resource for a Synchronous Line using Discrete Event Simulation
Restricted (Penn State Only)
- Author:
- Parasrampuria, Harshita
- Graduate Program:
- Industrial Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- March 17, 2024
- Committee Members:
- Russell Richard Barton, Thesis Advisor/Co-Advisor
Catherine Harmonosky, Committee Member
Steven Landry, Program Head/Chair - Keywords:
- Synchronous Lines
Discrete Event Simulation
Resource Scheduling
Downtime
Crane
Scheduling System
Statistical Analysis
Simio
Failures - Abstract:
- The research evaluates different scheduling policies for shared resources in synchronous manufacturing lines using Discrete Event Simulation (DES) to study various scenarios. It highlights the importance of smart scheduling of shared resources, like machinery and personnel, and examines the trade-offs between moving resources around versus keeping them stationary. This is crucial for improving operational efficiency and reducing downtime. There has been a lot of work done on scheduling in asynchronous lines, but synchronous environments, where parts move together, haven't been studied as much. This gap is what the thesis aims to fill by looking at simple scheduling strategies to better allocate resources, improve efficiency, and avoid bottlenecks. The study looks into five specific scheduling policies: First-In, First-Out (FIFO), Upstream Priority, Downstream Priority, Random Selection, and Round Robin. These are tested in both predictable and unpredictable settings, including the possibility of equipment failures. The research simulates real-life conditions to see how these policies would impact manufacturing efficiency and flexibility. It also starts the groundwork for creating a digital twin by building an isolated digital model. The potential for evolving this model into a digital twin through machine learning and deep neural networks is considered a future direction for this project. Through simulations and statistical analysis, the findings show clear differences in how well different policies work, with FIFO and Round Robin being particularly effective in both predictable and unpredictable conditions. This work contributes to the manufacturing field by providing solid evidence on which scheduling approaches work best under different circumstances and by offering useful insights for managing manufacturing operations.