Doctoral Dissertations

Alternative Title

Sensor data based decision making

Keywords and Phrases

Manufacturing sytem performance improvement

Abstract

"Increasing globalization and growing industrial system complexity has amplified the interest in the use of information provided by sensors as a means of improving overall manufacturing system performance and maintainability. However, utilization of sensors can only be effective if the real-time data can be integrated into the necessary business processes, such as production planning, scheduling and execution systems. This integration requires the development of intelligent decision making models that can effectively process the sensor data into information and suggest appropriate actions. To be able to improve the performance of a system, the health of the system also needs to be maintained. In many cases a single sensor type cannot provide sufficient information for complex decision making including diagnostics and prognostics of a system. Therefore, a combination of sensors should be used in an integrated manner in order to achieve desired performance levels. Sensor generated data need to be processed into information through the use of appropriate decision making models in order to improve overall performance. In this dissertation, which is presented as a collection of five journal papers, several reactive and proactive decision making models that utilize data from single and multi-sensor environments are developed. The first paper presents a testbed architecture for Auto-ID systems. An adaptive inventory management model which utilizes real-time RFID data is developed in the second paper. In the third paper, a complete hardware and inventory management solution, which involves the integration of RFID sensors into an extremely low temperature industrial freezer, is presented. The last two papers in the dissertation deal with diagnostic and prognostic decision making models in order to assure the healthy operation of a manufacturing system and its components. In the fourth paper a Mahalanobis-Taguchi System (MTS) based prognostics tool is developed and it is used to estimate the remaining useful life of rolling element bearings using data acquired from vibration sensors. In the final paper, an MTS based prognostics tool is developed for a centrifugal water pump, which fuses information from multiple types of sensors in order to take diagnostic and prognostics decisions for the pump and its components"--Abstract, page iv.

Advisor(s)

Saygin, Can
Sarangapani, Jagannathan, 1965-

Committee Member(s)

Ragsdell, K. M.
Grasman, Scott E. (Scott Erwin)
Dagli, Cihan H., 1949-

Department(s)

Engineering Management and Systems Engineering

Degree Name

Ph. D. in Engineering Management

Sponsor(s)

Boeing Company. Boeing Phantom Works
Missouri University of Science and Technology. Intelligent Maintenance Systems

Publisher

Missouri University of Science and Technology

Publication Date

Spring 2010

Journal article titles appearing in thesis/dissertation

  • Testbed architecture for auto-ID technologies
  • Adaptive inventory management using RFID data
  • RFID-based smart freezer
  • Mahalanobis Taguchi System (MTS) as a prognostics tool for rolling element bearing failures
  • Mahalanobis Taguchi system (MTS) as a multi-sensor based decision making prognostics tool for centrifugal pump failures

Pagination

xii, 173 pages

Note about bibliography

Includes bibliographical references.

Rights

© 2010 Ahmet Soylemezoglu, All rights reserved.

Document Type

Dissertation - Open Access

File Type

text

Language

English

Library of Congress Subject Headings

Inventory control -- Automation
Multiple criteria decision making
Radio frequency identification systems

Thesis Number

T 9646

Print OCLC #

747502981

Electronic OCLC #

747503855

Share

 
COinS