Good day, Dr. Sourav Kumar Mahapatra. I'm Arjun. I've had a look through your background and your professional record. I'm interested in hearing more about your work and how you see yourself fitting in here with our faculty. Ready to start?

Please start by introducing your professional journey in academia.

Good evening, my name is Sourav Kumar Mahapatra. I am from West Bengal and born and brought up in Odisha. I have completed my education from Odisha and currently I am working as a Project Associate at CSIR. I'm empty Bhubaneswar where I am working in an R&D project.

1 1/2 year of teaching experience as an assistant professor in Mechanical engineering and I have published 14 research papers in the field of computational modeling, composite materials, artificial intelligence and machine learning. I am looking forward for this role as uh, my research area is highly aligned with the role which is offered in the present UH.

Hi, Rol.

Good afternoon, Professor. Let’s dive into evaluating your experience and alignment with the role more thoroughly. First, could you elaborate on your approach to teaching computational modeling concepts? How do you ensure students grasp the fundamentals and effectively apply them?

I have I have worked in.

Different computational software such as ANSYS, Python, Machine learning and where I have I have also published first total 8 research papers and I I can teach Python as programming tool machine learning and I can also apply my computational skills to guide undergraduate, postgraduate and PSD research scholars so that they can do their work in research.

Related to computational modeling.

Understood. When teaching computational modeling using tools like Python or ANSYS, how do you balance the theoretical foundation with practical skill development? Could you provide a specific example of a teaching method or activity you’ve found effective?

Suppose uh, in, uh, in, in, in relation to uh, uh, experiment, which is related to students curriculum, such as, umm, finding the tensile load of a specimen or finding the flexural load of a specimen, which they will perform in their experimental, uh, in their experiments. So I'll just align them with the Ansys modeling so that so that they can take a feel how the. How the specimens are failing in tensile mode, in flexural mode in inter laminar shared strength mode?

Such that they will get an idea how the simulation part is going on and they can also match the results with the experimental 1. Moreover, they can also do the same programming using Python software, also using machine learning software to predict the material behavior in material modeling.

That's a thoughtful approach, Professor. By integrating experimental results with computational simulations and programming, you’re bridging the gap between theory and application. Now, could you elaborate on how you apply AI/ML in computational modeling, especially in materials science? Perhaps an example from your research publications?

Yes. So basically in my research, we are, uh, fabricating composite materials, uh, uh, and, and then we go for their characterization such as mechanical characterization, wear characterization, machining characterization. Whenever you perform a wear characterization, we need to perform a lot of tests to get the results, but because mechanical research is basically based on, uh, they, they, they depends a lot on.

Machines which are available to perform the experiments. But you can perform only a limited number of experience experiments in a particular machine, so it is very difficult to evaluate every parameter on each specimen. And because the machine is the machine, the number of experiments in the machine can be limited, so you can not actually evaluate.

The experimental results, So what we did is we performed some VR experiments and that is erosion. We are experiments from the composite material and we performed the experiments, got the data set and then we feed it in a machine learning model and then we predicted the output for for the erosion test. Similarly, what we did, we performed the machining characterization where we performed the drilling test from the drilling test. From the drilling experiments we found out the outputs such as the thrust force the top.

And the delamination factor and then we predicted that predicted the thrust force, torque and delamination factor and and then we matched with the experimental ones. Why we did this? We, we applied machine learning models because every time when we need a new data, we don't have to go and rush or do the experiments again. We can do it using the machine learning model that we have already feed and already. Made the machine learning model, so we'll just feed the data and we'll get the output.

I see—by leveraging AI/ML models, you're efficiently overcoming experimental limitations while gaining predictive capabilities. That’s a valuable application of computational modeling within materials science. How do you see this approach enriching the classroom experience, particularly when guiding research projects for students?

So we can use this AI and machine learning model in students project very effectively and efficiently. We can use it for undergraduate students, we can use it for postgraduate students, and we can do it as well for PhD students. What we'll do, we'll give a problem to them and we will tell them to perform the experiments on that particular area, for example.

To know the.

Can you know the mechanical properties of a composite material, for example, to know the, uh, to perform the mechanical machining test for a composite material. And then what we do, we take the those data model, we will train the students to develop their own machine learning model, to develop new machine learning model, to develop hybrid machine learning models, and then they will feed that model. They will train their own model and then they will test.

Their model they will make ready their model and then they can execute the model to ensure that it is working well and finally to make to ensure that their model is working good, they have to compare their experimental results with the results which they got from the models.

An excellent approach, Professor. Engaging students in both hands-on experimentation and modeling fosters a comprehensive understanding of both data-driven insights and practical validation. Let us discuss your proficiency in teaching theory and laboratory courses—how do you structure your classes, combining these two aspects, to ensure an impactful learning experience for students?

So, uh, basically when I teach students, I.

To use those concept on the application part we develop, we do some numericals so that they will get the feel that they can use their the same concept when to an experimental part or to an application part. So they can be clear both from the theoretical concept as well as from the experimental or numerical concept as well.

A balanced approach to theory and application, supplemented by visual aids and problem-solving, is commendable. Shifting to another critical area, how do you evaluate your students’ performances, especially for the projects or experimental courses? And how do you ensure fair and consistent assessment?

For project and experimental outcomes, first of all I'll give the I will divide my scrutiny into four parts. First part is experimental part, how the student is performing the experiments. Second part is how the how the results they got, whether it is good or bad or whether they are getting it incorrectly or not. Third thing is by conducting quizzes.

So that they can, uh, so that I, I will, I will get clear whether they are clear about the concept or not. And finally, I'll perform, uh, Vaibhav both to make, to get the insights from them or whether they have learned this, that subject or that experiment in a proper way or not.

And then I will evaluate whenever I'll give them uh, the uh, curriculum comes or whenever I'll give marks, I'll give equal marks to experiment, equal marks to the results they got, equal marks to quizzes, and equal, I'll give equal marks to the Viber box. So let's say if I'll, if I have 100 marks in a proper experiment, then 25 marks, 2 experiments, 25 marks to the results they got 25 marks to quizzes and 25 marks to vibes.

Your structured evaluation method demonstrates a clear commitment to fairness and a comprehensive assessment of both process and understanding. Now, let us discuss your experience in guiding research and student projects. How do you motivate students to pursue innovative research ideas and support them throughout their journey?

Uh, see as far as the motivation is concerned.

Uh, whenever, uh, I will teach undergraduate or postgraduate students, I'll motivate them to do research based, research based study because, uh, what research does is it makes you, uh, very clear about the concepts and it is very interesting, provided you give some time to research. So I'll start with undergraduate students. To read at least one paper in a month, and in the same manner I'll.

Motivate the postgraduate students to, uh, read a research paper in a week so that they can get an idea about their research and about, uh, what they have to pursue. That is #1 #2 is as soon as they will, uh, do this or they will read a paper, they will get an idea. And they will, I believe that they will, Some of the students will get interested to pursue research. And after that, when they will again come to me, I'll give them some problems.

And I will also tell them to do literature review on that particular problems, so that whether any research has already been published or not. After that, whenever all these things are done, I'll give them a research problem and we'll work upon it.

And finishing it will will try to publish a paper based on that research gap and will try to will make sure that the research will end up within in a publication or a patent.

Your layered approach, starting from introducing research papers to culminating in publications or patents, reflects a well-thought-out strategy for guiding students toward impactful academic contributions. Let’s now discuss your publication milestones. Could you share insights on one or two significant research papers that best showcase your expertise in computational modeling and its applications?

Yes. So I have 14 research publications.

If I'll, if I'll have to choose two papers, then one paper I'll choose from polymer composites that is Q1, Journal of Impact Factor 4.7 and that is the development of machine learning models to predict the wear behavior of polymer composites. What we did is we have developed polymer composites filled with Ongir and Slack and then we have performed the wear characterization tests. To to find out the. To find out the output, that is the erosion wear rate.

And then we have predicted the UH output using four different machine learning models, that is, decision tree, random forest, support vector machine, and gradient boosting machine. And finally, we found out the best model to predict the erosion to predict the erosion wear rate. And we found out that the gradient boosting model is the best model to predict the erosion wear rate of spongy and slack pill polymer composite.

Compare UH-2 composites which are made-up of two different techniques. 1 is vacuum assisted resin transfer molding technique and one is Handlab technique. What we did here is we perform different different mechanical tests such as tensile, flexural, interlaminal shear strength test, hardness test and impact test. And in the same way, we have also performed the same tests in a computational software that is ANSYS and we have compared each and every. Test with that of experimental results of VRTM technique and that of the hand layup technique and we found out that the computational results match well with the vacuum assisted resin transfer molding.

The 2nd paper I'll choose is a comparison of fabrication techniques to.

Results.

Both papers demonstrate strong alignment with computational modeling fundamentals and practical applications. The first highlights the integration of AI/ML in predicting composite wear behavior, while the second showcases your expertise in computational validation of experimental techniques. Now, shifting to a broader perspective—how have you contributed to or engaged with industry projects or consultancy roles, if any, as part of your academic or professional experience?

Right now I'm working on a, uh, research and development project which is funded by Ministry of Mines and the title of the project is development of lanthanum aluminium Titanic derived thermal barrier coating for aerospace propulsion. So basically what we did, because aerospace propulsion is a very broad area. And in that case we are designing turbine blades in which we give a coating on the turbine plates. Normally turbine plates are made-up of. Nickel super alloy, but that nickel. Super alloy is not sufficient to bear the high amount of heat which is coming and bombarding it. So what we do is we apply a thermal barrier coating which is made-up of lanthanum, aluminum, titanate. So basically we do the synthesis of that powder and using high velocity oxy fuel coat, high velocity oxy fuel thermal spray technique or by using suspension high velocity oxy fuel ther.

Develop a coating on that turbine blade and then it is being used by applying that coating. What it it will do is it will increase the IT will increase the thermal expansion coefficient of the turbine plate and it will also make it more heat resistant so that it it can it will be resistant to more thermal cycling. So that is the benefit which we get when.

The technique we umm.

Will use lanthanum, aluminum titanate. So this is the this is this is funded by the industry that is the Ministry of Mines. And right now I'm working on this project for past eight months and we have already developed some of the coatings and some of the, some of the powder synthesis process is still going on.

I see. Your work on thermal barrier coatings for aerospace propulsion highlights an impressive integration of materials science, computational modeling, and practical applications. It also illustrates your capacity to manage high-value research projects effectively. Let’s now discuss your PhD experience—what are the key takeaways from your PhD research, and how have they shaped your teaching and research approach?

So my PhD is. I have taken 4 years to complete my PhD, that is from August 2021 to. September 2025. My PSD is a very fruitful experience. I'm very lucky to get a supervisor, Professor Alok Sapati, who is my PhD supervisor. He has guided me very well and I have started my PhD research work from the from the first day when I got into got the admission into NIT Rahulkilla. So I have started my PhD work in the first year only.

I have published one paper in the second year I have published 2 papers, in the third year similarly 4 papers and the fourth year three papers. It is very exciting. I have enjoyed my PhD throughout my four years and through this four years we have also been, we are also being.

Used as teaching assistant to many practical courses and theory courses. So I I I I was teaching assistant for mechanical workshop for modern manufacturing process lab for.

I think in labs for engineering thermodynamics theory classes where we are guiding the B.Tech students and we have also, we have also demonstrated the different experiments to them and also have evaluated their performances in the lab as well as in the theory classes.

Your PhD journey reflects a stellar trajectory of consistent academic output and a balance of research and teaching experience. Publishing multiple papers while engaging as a teaching assistant showcases your dedication and capability in both research and education. Lastly, what do you think sets you apart as a candidate for this role, and how do you envision contributing to the long-term goals of our institution?

Sorry to interrupt, we're down to the last 5 minutes.

My career goal is to build a strong academic and research career in mechanical engineering with emphasis on advanced material surface engineering, thermal barrier coatings, composite materials, artificial intelligence and machine learning, computational modeling as well. And I wish that while working in vit, I'll get a strong emphasis which which which has a very strong emphasis on academic excellence and research productivity and industrial collaboration. I will be very.

Easily I'll I can fulfill my goals and moreover your second question.

Can I may I know your second question? Actually I forgot.

Certainly, Professor. My second question was: how do you envision contributing to the long-term goals of our institution, particularly in terms of teaching, research, and fostering industry-academia collaboration?

So I will be publishing more and more papers as, as far as I can. I, I can publish more and more papers, umm, and uh, and moreover, I will try to, uh, publish one patent every year. I'll try to get a funded project, industry funded project every year. Uh, which will enhance me and which will, umm, which will allow me to. Make BIT more proud and will.

Will integrate its academic excellence.

Thank you, Professor. Your emphasis on extensive scholarly output, pursuing patents, and securing industry-funded projects aligns well with our institution's objectives of fostering academic excellence and industry collaboration. That concludes the technical part of this interview. Is there anything else you would like to add or ask before we conclude?

No.

Thank you.

Yes.

Thank you, Professor. Just to confirm, would you like me to formally conclude the interview now?