Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Roychowdhury_Sayak_PhD_Dissertation.pdf (2 MB)

ETD Abstract Container

Abstract Header

Data-Driven Policies for Manufacturing Systems and Cyber Vulnerability Maintenance

Roychowdhury, Sayak
http://orcid.org/0000-0003-4306-9334
http://rave.ohiolink.edu/etdc/view?acc_num=osu1493905616531091

Abstract Details

2017, Doctor of Philosophy, Ohio State University, Industrial and Systems Engineering.
This research explores deterministic and stochastic policies to help organizations make data-driven optimal decisions. The two major application areas identified in this research are manufacturing and cyber security. In a recent report published by McKinsey Analytics, the manufacturing industry uses only 20%-30% of the potential of data analytics. This suggests that there are still plenty of opportunities to use analytics in manufacturing processes. In the first part of my research, I formulate an Integer Programming model for the “stamping” process in automotive manufacturing. I develop a production scheduling method for automotive stamping to maintain optimal inventory positions. In stamping, different types of parts are scheduled for processing in the press, which requires different die-sets to be mounted on the press. This has all the elements of conventional scheduling problems with tardiness objectives and setup costs. Yet, it also has capacity constraints and part production constraints. We show that these constraints make solution with branch and bound difficult for problem sizes of interest. In this research, I use the structure of the scheduling problem and implemented heuristic methods like Genetic Algorithm alongside Earliest Due-date (EDD) rules to prioritize production of parts with low inventory as well as minimize the number of die-set changeovers. I call this new method Genetic Algorithm with Generalized Earliest Due-date (GAGEDD). I illustrate the computational advantages compared with alternatives and show its benefits using data from a real life automotive stamping press scheduling problem to build a decision support tool for the schedulers. The second part of this research is motivated towards improving cyber vulnerability maintenance policies under uncertainty. A conservative estimate by McAfee in 2014 puts annual cost of cybercrime at US$375B. This is an important contemporary issue where role of data analytics and optimization have a lot to offer. Here I implement stochastic optimization procedures for cybersecurity applications, where learning is incorporated to account for future rewards. First, I formulate a Partially Observable Markov Decision Process (POMDP) model to derive policies for cases when the state of compromise of a host is uncertain. This method assumes there is no parametric uncertainty. Next, I implement Bayes Adaptive Markov Decision Process model (BAMDP) on a dataset obtained from the cyber logs of an organization using finite numbers of model scenarios. Earlier BAMDP formulations use infinite model scenarios. I also describe the benefits of using finite scenarios including the ability to solve the problem optimally as a POMDP. The resulting BAMDP formulation accounts for the parametric uncertainty caused by the lack of data for certain events. I use a point based value iteration method known as PERSEUS to solve both of these problems to generate a-vectors, that can be used to design optimal policies based on the belief-state of the system. Another benefit of using finite numbers of model scenarios relates to decision making for multiple identical systems, e.g., a “fleet” of identical Linux computer hosts. The issue of identical systems in machine learning has apparently received little attention despite the widespread relevance in data analytics. I propose a method for solving multiple identical system policy problems. The proposed method is based on a relatively large POMDP formulation with methods to compute the relevant transition, expected reward, and observation methods being provided. Then, I explore additional advantages of finite model scenario BAMDPs relating to the ability to incorporate reward-based or other learning in intuitive ways. Also, the speed of learning and the concept of “fast learning” and average learning time are proposed and explored computationally. In concluding, I offer suggestions about how this research can be extended to build more powerful models with faster learning capabilities to help decision makers.
Theodore T. Allen, PhD (Advisor)
Cathy H. Xia, PhD (Committee Member)
Gagan Agrawal, PhD (Committee Member)
118 p.
Industrial Engineering; Operations Research
Operations Research; Scheduling; Automotive Manufacturing; Stamping; Genetic Algorithm; Partially Observable Markov Decision Process; Bayesian Adaptive Markov Decision Process; Cyber security; Cyber vulnerability maintenance;

Recommended Citations

Citations

  • Roychowdhury, S. (2017). Data-Driven Policies for Manufacturing Systems and Cyber Vulnerability Maintenance [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1493905616531091

    APA Style (7th edition)

  • Roychowdhury, Sayak. Data-Driven Policies for Manufacturing Systems and Cyber Vulnerability Maintenance. 2017. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1493905616531091.

    MLA Style (8th edition)

  • Roychowdhury, Sayak. "Data-Driven Policies for Manufacturing Systems and Cyber Vulnerability Maintenance." Doctoral dissertation, Ohio State University, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=osu1493905616531091

    Chicago Manual of Style (17th edition)

osu1493905616531091
280
© 2017, some rights reserved.Creative Commons License Data-Driven Policies for Manufacturing Systems and Cyber Vulnerability Maintenance by Sayak Roychowdhury is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by The Ohio State University and OhioLINK.