Heramb Narendra Somthankar

Highly motivated and skilled Predictive Maintenance Engineer with a strong background in Python programming and machine learning, seeking a challenging role as an Engineer - Data Analytics, Predictive Maintenace, AI Engineer, Data Scientist Condition monitoring, etc. Eager to leverage my expertise in Python libraries like Numpy, Pandas, and Matplotlib, along with my proficiency in implementing machine learning algorithms using sklearn and TensorFlow, to contribute effectively in managing and developing IOT data in the Google Cloud platform. With a passion for data research and domain knowledge in equipment controls, mechanical, and electrical functionality, my objective is to drive innovation and provide meaningful insights through data analysis, while fostering a diverse and inclusive work environment reflective of the company's values. I am excited about the opportunity to relocate to any location in India and work in a collaborative setting that values expertise, productivity, and efficiency, ultimately contributing to the company's mission of making life better for employees, customers, and communities worldwide.

During my tenure as a Master Thesis Student at Bayerische Motoren Werke AG (BMW), Dingolfing, Germany, I focused on optimizing industrial maintenance using Digital Twin technology. My thesis project involved validating Digital Twin models for hairpin production stations, aiming to provide reliable predictive maintenance solutions by evaluating the influence of gripper behavior on production. Through a comparative analysis of sensor data from real machines with DT simulation results, I assessed the accuracy and effectiveness of the DT model in predicting the effects of wear on stator hairpin production.

As an Intern Industrie 4.0 at BMW, I worked on preprocessing data from cutting stations (hairpin stators) using Python tools like Scipy, Numpy, and Pandas to predict the remaining useful life (RUL) of cutting tools. I developed a range of machine learning models, employing ScikitLearn and Deep learning techniques, with hyperparameter tuning to facilitate predictive maintenance of cutting tools. Additionally, I utilized the Hardware in Loop (HiL) technique to optimize digital twin simulation, extracting raw data from cutting machines and improving the existing simulation model. To enhance accuracy and efficiency, I implemented ISG-Virtuous and Beckhoff PLC to generate synthetic data, enriching the HiL simulation process.

Previously, during my role as an IoT and Machine Learning Intern at Cognifront, Nashik, India, I provided guidance to students on IoT projects and supported colleagues in implementing Machine Learning and Deep Learning algorithms. I also contributed to creating PowerPoint presentations and documentation in the areas of IoT, Machine Learning, and Deep Learning.

During my ongoing M.Sc. in Networked Production Engineering at RWTH Aachen University, I have gained comprehensive knowledge in essential areas that are highly relevant to the current industry trends. The program has provided me with in-depth insights into Industry 4.0 and Embedded Systems, enabling me to understand the integration of smart technologies and digitalization in modern production processes. Additionally, my coursework in Artificial Intelligence and Data Analysis has equipped me with the skills to leverage advanced algorithms and techniques to derive valuable insights from large datasets. Moreover, I have acquired expertise in Quality Management and Industrial Monitoring of Engineering Systems, which has honed my ability to ensure and enhance the efficiency and effectiveness of production operations. This well-rounded education has prepared me to excel in the field of Networked Production Engineering and contribute significantly to innovative solutions in the industry.