Good day, R. Prabhu. 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?

Yeah. Good day. Thank you.

My name is Prabhu. I did maybe in mechatronics, me in computer integrated manufacturing and I I did my BPS to work in the area of artificial intelligence in manufacturing. I have I have solved problems in abrasive water jet machine that is first tooth problems. I have solved first thing is in abrasive. What is it milling process prediction of process parameter is very.

Therefore, to avoid that, I have developed machine learning algorithms, uh, such as support regression, support vector regression, artificial neural network, decision tree, random forest. I have developed this model and I have predicted the responses like, uh, surface roughness, milling depth curve angle. OK, I, I have based on this machine learning, all the algorithm I have published, uh, repeated journal with an high with a high impact factor of 7.56 point.

Uh, very tough for manufacturing engineers, uh, the in industry they are practicing trial under method, uh, that uh, results in increase in production cost.

Uh next Internet is International Journal of uh, uh advanced manufacturing technology with an impact of uh 3.5 and one more uh journal of manufacturing processes I have published with an impact feature of 6.8. And a second problem is in abrasive waters in milling. Uh cutting head follows A raster path machine cycle. During the raster path machine cycle, the the cutting head undergoes acceleration and deacceleration phase.

Expert system with applications with an impact factor of 7.5 and then.

So during the deacceleration phase, uh, there is uh, overration occurs on the workplace network in general. Uhm, in manufacturing engineers are practicing mask method that results in wastage of mask material, wastage of machine resources, machine resources and then, uh, it increases production time also. So in order to avoid that, I have adaptively controlled the process parameter whenever the cutting head is slowing down.

I have adaptively reduced the flow of abrasive particles so that over resonance erosion will avoid. So I have developed a mechatronic system and implemented in abrasive origin machine and I have filed a pattern and I have received Fer and I also filed A Fer also and I have converted that pattern work into UH.

Pepperoni salt OK that that is published in journal of manufacturing process with an impact factor of 6.8 in my in in a pod I have I have attended a lot of workshops. I have I have had a lot of courses in Python machine learning in Python programming. So I have a good knowledge in developing machine learning algorithm and implementing that machine learning algorithm in the.

Urgent machine or any non traditional machines like additive manufacturing, uh, using Raspberry Pi 5 sensors and I have controlled numbers in case of mission, I have controlled the process parameter also. So I, I have have I am, I'm having good knowledge in implementing this artificial intelligence in machine mission. I mean non traditional machines, uh, that could be a Bridgewater machine or any non traditional machine, not only a Bridgewater jet machine, I can also implement.

Add into manufacturing also in additive manufacturing, uh, for example, uh, if, uh, United machine the IT, it is uh, adding the layer by layer. So it results in increase in surface roughness in this after completing the product, there is increase in surface softness. So I can also use this miscellaneous algorithm to predict the process parameter to reduce that surface roughness so that it increases fatigue life also.

So I can apply my machine learning algorithms to predict that process parameter in additive manufacturing process such as like laser powder based fusion like that I can apply my knowledge in that round machining processes also. So I have I have had a good knowledge in implementing cyber physical system in machining process.

Understood, Professor. Your insights on using machine learning for adaptive control in non-traditional machining processes like abrasive water jet milling and additive manufacturing are compelling. Additionally, your patent work and practical applications of cyber-physical systems reflect significant innovation. Let’s now verify your expertise in a structured manner aligned with this position. First, could you succinctly summarize how computational modeling was utilized in developing the adaptive control system you described? Specifically, focus on the modeling approach and tools implemented.

OK OK I have used the computational modeling for example during the let me take that whenever cutting head is moving in raster path cycle it it goes fast in travel travel in stepo region it will slow down again a fast in the travel cylinder region again slow down. Whenever doing that fast movement that the X axis motor pulley it will rotate in high fast way. When it is reaching the stepper region X axis stops.

And this Y axis starts rotate so that it will complete Step 4 region. OK I will fix. I have fixed the sensor in the X axis multiply using a encoder whenever the.

X axis motor police is slowing down. I have picked using encoder. I have picked that acceleration and deacceleration phase of the cutting head. Based on that I have developed the machine learning algorithm. Using uh Raspberry Pi 5I have used uh that is a support regression and I have.

Artificial Neural network and Decision Random forest K means clustering where whenever it is reaching that position. In that algorithm I have written that it need to reduce the flow of abrasive particles. So I have implemented that algorithm in Raspberry Pi so that Raspberry Pi it will automatically control the.

Abrasive flow using that actuator, pneumatic actuator using that Raspberry Pi OK that is here I have I have collected more number of data that is from X axis Motorola as well as from Oaxis Motorola. Based on that experiment data set I have developed machine learning algorithm. Whenever acceleration is that is cutting it is slowing down it will it will adaptively reduce the abrasive flow.

So that it will avoid overrun on the workpiece material. So I can also implement the same concept in any non traditional machining process not only abrasive machine. Once I have a knowledge in implementing real-time real-time control of process parameter in one machine then I can I could able to control any process parameter. Whatever I needed I can modify the. That is algorithm based on the input features I can I can activate the.

Corresponding process parameter so.

I I have had a good knowledge in uh, implementing uh, a that is uh, uh, in non traditional missioning process.

Thank you, Professor. Your explanation provides clear evidence of your ability to effectively apply computational modeling techniques, as well as machine learning algorithms, for real-time control in machining processes. Your work demonstrates expertise in leveraging data, neural networks, regression, clustering, and Raspberry Pi for adaptive control in industrial settings. This aligns well with the innovative use of computation in manufacturing. Let us now transition to discussing another pivotal aspect of the role. Could you discuss your approach to adapting AI/ML techniques to materials science problems, particularly in areas such as predicting properties or optimizing manufacturing processes?

Yes, for example in case of additive manufacturing itself if in that the FM that is for example laser powder.

Base. A fusion in that uh laser, it will pause on the in corner powder or titanium powder so that it will uh it will it will fuse the uh powder that is internal titanium powder after that again this comes down again that flow particles it will uh it will fill on the Bay and the top of the bed again that laser process it will move and it will fuse the material OK, if I'm controlling the intensity of laser and.

And the speed of curing if, if if I'm controlling this parameter, that material characterization that is a inside that, that is a formation of uh, uh, material reactions, it will changes if I'm changing the movement of laser and grain size and temperature. If I'm changing the temperature and based on the input that is angular. So orientation also by by changing this input parameter, how the material?

Of this laser path and powder dimension and.

Changing therefore I will I will contact experiments you by by by changing the input process parameter values after after based on some design of expert and I will conduct experiment and I will conduct. After conducting experiment I will go I will go to cinema analysis and how it is changes based on changing the input parameter by having the data set, I will build misleading model on the data set I will build. Model on the data set so and since misleading model algorithm has a high, it can it can have the complex relations which will find the complex relationship between the input and the output. So that if that mission algorithm will have the that is the relationship between how it is behaving based on input and output so that I can predict the. That is a metal characteristics based on the input or after developing the model.

I can check the after building model I will predict the process parameter like metal characterization and again I will value valid validate that one by in real time. Whether it is model is giving the exact thing as that by in real time I will validate that one. Likewise I can implement this computational modeling.

Umm, in additive manufacturing for material characterism.

Thank you, Professor. Your ability to articulate the use of computational modeling and machine learning in improving the material characterization process in additive manufacturing is commendable. Your structured approach—designing experiments, analyzing microstructural changes under varying input parameters, modeling complex relationships, and validating model predictions in real-world scenarios—demonstrates a strong foundation in both theoretical and practical aspects of materials science and AI/ML integration. Now, let's move to your proficiency in teaching both theory and laboratory courses effectively. Could you outline your approach to creating an engaging and structured learning experience for students in a computational modeling course? How do you ensure that the students grasp both the theoretical concepts and their practical applications?

OK, thank you. OK, I will. I will First of all in I will explain the machine learning algorithm using Python. I will explain the theoretical concept clearly in the classroom. I will Python that machine learning algorithm, it is in Python. It is fully based on OOPS concept that is object oriented programming. I will explain the basic concept of OOPS that is objective programming and the algorithms using that Python program. And whoops concept and how to extract the features using that Python And how to build the model using the data set and how to fix the OR how to fit the model on the data set. I will explain each and every element of Python programming and machine learning algorithm. I will explain the basics Max behind that machine learning algorithm and how it is beha behaving like for example supported regression how it is behaving and if it is K means clustering how?

I will make them to understand theoretically how it is, uh, working once if they're thorough with the mathematical equations of, uh, machine learning algorithm, they will have some good idea. OK, this, this is the way that algorithm is working. So that I will make them to understand and realize the mathematical background of missionary algorithm by solving numerical, uh, problems relative to that, uh, mathematical, I mean machine learning algorithms. So after, after I will teach that one in the classroom.

This, uh, clustering the data set based on the relationship like, uh, within cluster, uh, sum of square. I will explain that how that Euclidean distance is, uh, calculated. And I will explain by the help of case study mathematical discussion. I will make them to understand theoretically.

OK, how about my music? I will also the I will also give, uh, case study. I will also simulate, I will also use, uh, uh, Python, Anaconda, uh, uh, using that Jupyter notebook. I will make them to run that coding machine learning algorithm. What about the predicting accuracy? I will make them. I will practice them. I will give hands-on training to them so that they will easily understand, uh, the. Basic thing and I will I will implement the same thing in the laboratory also I will implement that.

Uh, visual learning algorithm, uh, using Raspberry Pi 5 sensors, activators in any one of the machines once if I make them to understand in one machine in laboratory that can easily understand the concept, whatever they have learned in theory, they can learn in the lab.

Practically they can learn that OK this is how it is controlling like that. I will make them to understand a practical way also by implementing sensors, Raspberry Pi and activators in the one of the machines in the laboratory. For example abrasive autogen machine are added to manufacturing machine. Any machines I will control the process parameter using sensor as verify machine learning model and then actuator. If I whatever I have taught in the class, I will make them to understand in the lab so that if they if they are asking any doubt also I will clarify in the.

A practical thing in the lab itself, they, so that they could understand, OK, this machine algorithm behaving like this and it is controlling like this, like then I will, uh, I will make them to understand in a practical way in lab also. So I will, I will, I will make the two thing in this simultaneous that this is a, I will make this a synchronous learning in both in, uh, lab as well as in theory classes. Whatever doubt they're asking, I will explain in the both in class as well as in. Uh, laboratory so that, uh, they can, uh, uh, have a very good practical knowledge so that they can implement their innovative idea. Uh, using that my, I have ideas, whatever, whatever I have thought to them, they can implement their own innovative idea and then they can make a lot of, uh, uh, beautiful things in, uh, uh, technologies in a, in manufacturing, They can make a lot of beautiful things in the, uh, in manufacturing.

Understood, Professor. Your approach to fostering a combination of theoretical understanding and hands-on laboratory experience is well-structured and balanced. Emphasizing fundamentals like object-oriented programming, mathematical problem-solving in machine learning, and practical application using tools like Jupyter Notebook or physical systems such as Raspberry Pi and sensors shows a strong commitment to structured teaching. It also highlights your ability to bridge theoretical depth with practical implementation, a key quality for creating a fertile learning environment. Now, could you elaborate on how you evaluate student performance effectively, both in terms of theoretical grasp and practical application, ensuring overall academic growth?

Yes, Sir. Thank you. OK, I I will evaluate student in theoretical way by giving assignment or by giving many projects so that they can implement their idea using that projects and case studies. And I will I will give the journals also I will I will make them to study the literature survey what others have done. And how they have solved that problem so that theoretically they will, uh, get more idea?

In, uh, practical way, I will, uh, I will make them to fix the sensor and the machines, uh, like for example, missions to monitor the Orphas also, I will make them to fix the sensor and I will, I will make them to get the data from the sensor. And I will make them to, uh, develop model and, uh, use model and fix that model in the Ras 355 and I will make them to control. Any process parameters so that I I will evaluate their?

By the help of case study. By studying the case study.

For example, in label, so I will evaluate perfectly whether they are doing I will go and monitor them whether they are really doing the work or not. Practically, I will ensure that they have to be encased in the in doing that work really in in machines they need to I will monitor their report also whether they're doing correct or not and I will record note also I will maintain them. I will give good marks only to to the persons who is doing work and if they are not doing.

I will, I will not, I will make them to understand that concept. But why you're not understanding that one? Uh, do you have any doubt in that? So that I will clarify that and I will make them to do that work in practical way in lab also. So in that aspect, I will evaluate the students by continuously monitoring their performance by monitoring in every aspect, that is by how they are getting the results. So I will in, in case of theory, I will give case studies, literary survey indicate of in.

Of laboratory, I will continuously monitor them and if they're they're having any doubts, I will explain that one and I will make to understand that concept in a practical way.

Thank you, Professor. Your emphasis on continuous monitoring, detailed evaluations through assignments, case studies, and projects, combined with your focus on real-world applications in the laboratory, demonstrates a well-rounded and rigorous approach to student assessment. By blending theoretical and practical evaluations alongside personalized mentoring for students struggling with concepts, your method ensures holistic academic and professional development. Next, could you share your experience in guiding research projects or publications and how you encourage students to contribute to scholarly outputs, such as research papers or patents?

Yes, I have published high impact part of journals in expert system with with an impact partners on .5 by by and IJMT that is International Journal of Advanced Manufacturing technology with the impact of 3.5 and one more paper Journal of Manufacturing processes. I have used machinery algorithm like support with regression neural networks. Resistant to random forests. OK, I will. I will guide them.

How to write a paper, how to write a technical paper to the students, How to find the novelty in the work based on the novelty. How we are to work to fill that gap in the research gap. And what are the things we have to do And how to write a paper in a good way to fill to, to answer that novelty, that is to to solve that novelty concept. And I will make them to work relative to the novelty. Experiment. I will conduct experimentation how the how we are getting the results. Cells whether the novelty, whether the student is getting the novelty from the back, I will ensure them and I will keep on guide them in a technical aspect by why it is happening and why it is not happening. And I will find physics from their work and I will make them to understand physics what they are, what they are understanding from their back. And I will make them to understand so that I will make them to involve in that back so that I will convert. I will also get the them to convert.

That work into international journals and so that they will also get confident in the corresponding technical aspect and they can also prove that paper also will support them while they're going for interview also. So that I will be supporting every aspect to make them to get to become a good technical engineer.

Yes, thank you. Actually in abrasive orders in machine that did, I'm cutting here that RFS gets wear over the period of time in manufacturing. I have, I have gone through the several manufacturing industries and all using abrasive water admissions. I have Venkat, what is their problem And they said that RF is getting weird, we are not able to find out. So that is the major problems. Then I have took that problem as a as a researcher, Yeah, I mean research. The problem and I told I have found that our face wear whenever there is a change in wear there is a drop in water pressure so that I have fixed that water pressure. So I calibrated that change in Rfis and I converted that using that pressure sensor and I can I have connected the tone to IoT and I using machine learning algorithms I have predicted the orifice pair so.

Thank you, Professor. Your track record in publishing high-impact research, combined with your hands-on approach to mentoring students through the process of identifying research gaps, designing experiments, and understanding the underlying physics, demonstrates a deep commitment to fostering their academic growth. Encouraging students to publish their work in reputable journals while emphasizing the relevance of such achievements in professional settings underscores your dedication to their success both as researchers and as future engineers. Let us now move to one last critical aspect. Could you detail your experience, if any, in industry-related projects or consultancy work, and how that experience informs your teaching or research activities?

I have implemented that one. Now I am going to convert that one to pattern and I am going to publish in good journal also.

Thank you. Thank you, Ernest. Thank you. Thank you, thank you.

Professor, to begin, could you share which computational modeling course you've found most engaging to teach, and what specifically makes it rewarding or challenging for you?