Thesis Defense - Kenan Cem Demirel (MSIE)

Kenan Cem Demirel - M.Sc. Industrial Engineering

Asst. Prof. Erinç Albey – Advisor

Date: 18.01.2021

Time: 09:00

Location: This meeting will be held ONLINE. Please send an e-mail to gizem.bakir@ozyegin.edu.tr in order to participate in this defense.

An Application of Digital Transformation: Predictive Maintenance Scheduling

Thesis Committee:

Asst. Prof. Erinç Albey, Özyeğin University

Assoc. Prof. Okan Örsan Özener, Özyeğin University

Assoc. Prof. M. Güray Güler, Yıldız Technical University

Abstract:

Current technological developments in the industry and literature with the fourth revolution of industry make it possible to collect accurate data from production processes and develop data-based models to increase production productivity. The management decisions can be made more intelligent and automated with process tracking systems and predictive models. Predictive Maintenance (PdM) policies stand out in this area as an opportunity, especially for mass production facilities. Necessary early actions can be taken and maintenance activities can be planned by monitoring equipment conditions and estimating failure times. At this point, creating maintenance schedules for extensive facilities using predictive model outputs also emerges as another problem.

Within the scope of this thesis, an end-to-end digital transformation application has been carried out as a pilot study on a mass production line of an international scale production facility. A data collection infrastructure is established to collect process data from sensors and inspection team feedback about the equipment conditions. A PdM approach is introduced to estimate indicator scores about the equipment conditions and remaining useful lifetimes (RUL) by using the collected data. As maintenance activities require long-term operations, scheduling these activities is considered as a multi-campaign (shutdown) scheduling problem with sequence dependent maintenance durations. A mixed-integer linear programming (MILP) model is developed, and the outputs obtained from the predictive model are fed as deterministic inputs into the model to schedule maintenance activities. Open-source applications are used in the developed solutions; thus, continuous improvement and sustainability with lower costs are aimed.

Bio: 

Kenan Cem Demirel received his B.S. degree in Industrial Engineering from Özyeğin University in June 2017. He started his business life before graduation and worked in departments such as supply chain, customer analytics, and risk assurance in multinational companies for about three years. In 2018, he joined the Master of Science program in Industrial Engineering at Özyeğin University and has been working under the supervision of Asst. Prof. Erinç Albey. His research focuses on predictive models, industry and customer analytics. He is currently working as a Data Science Engineer at Özyeğin University in the Department of Information, Technology Transfer and Entrepreneurship.