Team members:
- Nithursika Kalanantharasan
- Sangaran Thevarasa
- Pairavi Thanancheyan
This competition featured three main rounds:
- Storming Round
- Semifinal Round
- Final Round
In each of these rounds, our team developed an end-to-end machine learning pipeline to tackle the challenges presented. We leveraged advanced analytics techniques to solve complex problems and make data-driven decisions.
Please explore the relevant folders and documentation within this repository to learn more about our approaches, methodologies, and the code we used during the competition.
We hope you find our work insightful and informative. Feel free to reach out to us with any questions or feedback.
The Storming Round of the competition focused on store profiling. Store profiling is a crucial analytical process used to analyze the performance of stores based on their sales and customer behavior. The primary objectives of store profiling are:
- Identify stores that are performing well.
- Identify stores that are underperforming.
- Optimize the process of item range decision.
In this round, our team leveraged advanced analytics and machine learning techniques to address these challenges. We developed data-driven solutions to help businesses make informed decisions about store performance, resource allocation, equipment, marketing, and staffing.
In the Semifinal Round of DataStorm 4.0, we were tasked with supporting Beverages Company XYZ, with whom we had previously provided an analytics solution during the Storming Round. This time, Company XYZ sought our expertise to enhance their asset allocation process through advanced analytics.
The key objectives for the Semifinal Round were as follows:
- Enhance asset allocation processes for Company XYZ.
- Utilize advanced analytics techniques to optimize asset allocation decisions.
- Build upon our previous analytics solution to provide even more valuable insights.
Our team rose to the challenge by developing advanced analytics models and strategies to help Company XYZ make data-driven asset allocation decisions. We leveraged our previous work and incorporated additional methodologies to further enhance the decision-making process
In the Final Round of DataStorm 4.0, our challenge was to optimize product assortment and volume allocation within the allocated freezer types for Beverages Company XYZ. Following our successful provision of an analytical solution in the Semifinal Round, Company XYZ relied on our expertise to maximize their total margin earned from each of their stores.
The key objectives for the Final Round were as follows:
- Optimize freezer assortment and volume allocation for each product.
- Maximize the total margin earned from each store.
- Ensure a balanced product assortment within each segment.
- Avoid stock shortages and maintain efficient order fulfillment.
Company XYZ allocated freezers to their stores based on our previous recommendations from the Semifinal Round. In this round, we leveraged historical sales data and implemented advanced analytics solutions to address these objectives. We optimized the allocation of products to the recommended freezer types while adhering to constraints such as a maximum limit of 30% of the recommended freezer type's volume for any single product and ensuring that the total stock distributed is below the current available stock.