I am Srikanth, a data enthusiast and an Industrial Engineer by degree, Strategic Projects Manager by profession. A friend of mine introduced me to Great Learning, and that was how I came to know of this program.
Before my PGP AIML program, I was not aware of how to analyze the data and clueless about the statistics used for the analysis, and I was not satisfied with my contribution to what I was delivering to my organization and to my career. I had no idea about programming or even the basic concepts related to AI/ML before my PGP program.
This program gave me the zeal to learn more about this field and to keep me on par with my peers. I still am learning and will continue to do so all my life. I have been able to apply what I have learned from this program to my work. My current workplace deals with some supply chain-related challenges. 75% of the cost of the organization comes from the supply chain function. I have used the Ensemble Technique concepts that I learned from my PGP course. I predicted the cost drivers well in advance & performed the projects effectively. Almost all the models that we create require data gathering and cleaning, and to do that to its fullest, I need to analyze and understand & interpret the data I have in hand. The decision models learned from my course are being used in an organization.
This helped in saving supply chain costs to a greater extent. The problem statement on one of the projects in which I’ve applied the ensemble technique was “ Establish a Forecasting model on the Supplier Excess delivery.” The supplier agreed to send the raw materials as per the negotiated tolerance, up to 5 % more than that actual demand. Because of not being able to predict how much the supplier can send at the beginning results in raw material leftovers which are worth the value of one million USD per year write-offs.
A detailed project plan was made with the data-gathering plan & collected the data for the past 5 years to understand the supplier behavior. Imported the data in python, performed the feature engineering, performed exploratory data analysis, split the data with train and test & built the ensemble technique. Performed hyperparameter tuning and pickled the random forest algorithm as the best-evaluated model. With the above methods, we are now able to predict the supplier behavior at the beginning and have saved more than 500,000 USD in the same year.
Learning advanced technology with data-centric deep study can put us at the top of the race & help in contribution & growth in our career.
