Date of Award

2024-08-01

Degree Name

Master of Science

Department

Industrial Engineering

Advisor(s)

Tzu-Liang (Bill) Tseng

Second Advisor

Ivan A. Renteria Marquez

Abstract

Industrial developments over the past few decades have transformed manufacturingprocesses, emphasizing the necessity for efficient scheduling systems. The shift from manual scheduling to automated and optimized systems has underscored the need for innovative approaches to improve production efficiency. Scheduling optimization of manufacturing systems is critical to improve operational efficiency. Typically, these scheduling problems become challenging because they involve coordinating various production tasks, resource management, and makespan minimization. These challenges are particularly evident in flow-shop manufacturing systems, where the operation sequencing significantly impacts overall performance. The main objective of this manuscript is to present the methodology followed to optimize the schedule of a flow-shop manufacturing system through the combination of discrete-event simulation and neural networks. With a specific focus on enhancing operational efficiency, the study aimed to employ the built-in neural network capabilities of Simio simulation software to optimize the makespan, consequently reducing the overall production time required for processing manufacturing orders. By exploring the integration of advanced simulation techniques with neural network functionalities, this research highlights the advantages of this powerful tool to model and optimize complex manufacturing processes. Moreover, the study compared the Simio with neural networks approach and Palmer Heuristic method to assess their respective advantages and disadvantages and emphasized the potential benefits for the manufacturing industry by implementing these methodologies to facilitate informed decision-making and drive improvements in productivity and resource utilization within the context of manufacturing operations.

The findings from this research offer valuable insights into the effectiveness of combiningdiscrete-event simulation with neural networks, providing a framework for future studies and practical applications in manufacturing systems. The demonstrated improvements in makespan and overall operational efficiency highlight the potential for these advanced techniques to be adopted in different industrial settings.

Language

en

Provenance

Received from ProQuest

File Size

57 p.

File Format

application/pdf

Rights Holder

Jesus Ricardo Herrera Garfio

Download

Included in

Engineering Commons

Share

COinS