* * * * *

Kerelous Waghen

**** ****. *******-*********, # ***, Montréal, QC, CA H3T 1J4

Cell phone: +1-514-***-****

Email: add1k4@r.postjobfree.com

LinkedIn: https://www.linkedin.com/in/kerelous-waghen-35253a174/ Research gate: https://www.researchgate.net/profile/Kerelous_Waghen Kerelous is an AI research scientist on the reliable application of artificial intelligence (AI) in industrial systems for optimizing the system performance and manage its associated risks. His main area of expertise is in data science, machine learning, data engineering, maintenance and reliability engineering. Kerelous has in-depth experience in the explainable AI (XAI), intelligent decision support system (IDSS) and causality analysis in complex systems. During his Ph.D., he proposed a novel interpretable data-driven graphical trees that reinforce the application of XAI in the industrial systems. The proposed conceptual methodology named interpretable logic tree analysis (ILTA), which address the challenges for modelling the causality structure and its changing over time for complex events. ILTA methodology able to automatically construct a fault tree analysis (FTA) model based only on the extracted knowledge from system databases. Two advanced trees are proposed based on ILTA as the following; MILTA that able to graphically decompose the complex event such as fault to its hierarchical causality structure in form of a graphical multi-level tree for addressing the challenges of the fault diagnosis task. While ITCA addressed the challenges of the fault prognosis task in form of a graphical time multi-level tree for modelling the changing in the complex fault hierarchical causality structure due to system degradation and system ageing. EDUCATION / PROFESSIONAL DEVELOPMENT

Doctor of Philosophy (Ph.D.) in Industrial Engineering Jan. 2017 – Feb. 2020 Department of Industrial Engineering and Applied Mathematics, Polytechnique Montréal, Québec, Canada. Research-based Master (MSc) in Industrial Engineering (change the program to Ph.D.) Jan. 2016 - Dec. 2016 Department of Industrial Engineering and Applied Mathematics, Polytechnique Montréal, Québec, Canada. Bachelor of Science (B.S.) in biomedical Engineering Sep. 2005 - May 2010 Higher Institute of Engineering, El Shorouk Academy, Cairo, Egypt. PROFESSIONAL EXPERIENCE

CanmetENERGY - Canada (Ph.D. Researcher) From May. 2018 to Apr. 2019 Kerelous was working in the Program of Energy Research and Development (PERD) that its research objective is ensuring a sustainable energy future for Canada. During this period, he worked on the following two projects;

• He studies the significant impact of the multiple system KPIs on explaining the causality structure from different perspectives for a given fault related to the excessive energy consumption in the evaporator system based on his proposed Ph.D. methodology.

• He developed a solution based on his proposed Ph.D. methodology that discovers and defines the multi- performance states in complex systems based on his proposed Ph.D. methodology. NMC Healthcare - UAE (Sales engineer) From Jan. 2014 to Nov. 2015 Kerelous was working as a sales engineer, his main tasks are to prepare and deliver technical presentations explaining products or services to customers in the healthcare industry domain. Orion Medical Systems - Egypt (co-founders) From Aug. 2011 to Dec. 2013 kerelous was one of the co-founders of Orion medical systems, he led the design and R&D team that their main tasks were developing new products and improving the performance reliability of the existing products. 2 P a g e

Egyptian Biomedical Engineering - Egypt (Field Service Engineer) From Aug. 2010 to Sep. 2011 Kerelous was working as a field service engineer, he was one of the maintenance team that its task was performing the frequency and urgent maintenance for the medical equipment. PROJECTS AND TECHNICAL SKILLS

Projects

• Causality analysis based on XAI for understanding the differences between the performance states

causality structure in the industrial actuator

system.

• Deep neural network (DNN) for predicting the

different performance levels of CPU processors.

• Time causality analysis based on XAI for

understanding the effect of ageing on changing

the causality structure for NASA turbofan engine.

• Recurrent neural network (RNN) for forecasting

the machine remaining useful time and optimizing

maintenance decision making.

• Understanding the customer's behaviour and

their conditions based on XAI for building a credit risk model.

• Automated fault detection and classification

based on restricted Boltzmann machine (RBM)

and softmax regression

• Discovering the possible breast cancers causality based on XAI for assisting cancer research

medicine.

• XGBoost model for forecasting the changing in

house prices and extract the related important

feature.

Technical Skills

Languages: Python, R, C, Pig, Hive, SQL.

Platforms: IBM Cloud, IBM Watson studio, Azure, Hadoop. Tools: STATISTICA, Matlab, Explore, Weka, PowerBI. ATTENDED COURSES AND CERTIFICATES

Graduate Courses

• System Engineering Maintenance Management • Programmation en nombres entiers

• Fiabilité des actifs en exploitation • Ergonomie occupationnelle : aspects physiques

• Méthodes statistiques d'apprentissage • Atelier de formation en santé sécurité

• Analyse de régression et analyse de variance • Séminaires de génie industriel

• Data mining techniques • French courses at Université de Montréal Certificates

• IBM, Big Data Foundations, Level 1 and 2 • IBM, Spark

• IBM, Hadoop Foundations • IBM, Cloud essentials

PUBLICATIONS

Waghen, K., & Ouali, M. S. (2019). Interpretable Logic Tree Analysis: A Data-Driven Fault Tree Methodology for Causality Analysis. Expert Systems with Applications. Waghen, K., & Ouali, M. S. (2019). Multi-level Interpretable Logic Tree Analysis: A Data-Driven Approach for Hierarchical Causality Analysis. (submitted paper). Waghen, K., et al. (2019). A hybrid framework for defining the system multi-states based on MILTA data-driven causality analysis and expert knowledge. (submitted paper). Waghen, K., & Ouali, M. S. A Data-Driven Fault Tree for a Time Causality Analysis in Aging Systems. (submitted paper).

K. Waghen et al. (2019). A novel data-driven fault tree construction methodology for supervised industrial data labelling approach. (CanmetEnergy internal report).

Contact this candidate