.webp&w=3840&q=100)
Industry
retail-and-cpg
Skills
data-understanding
approach
data-wrangling
data-visualization
data-quality
problem-understanding
ml-modelling
quality
code-versioning
git-version-control
programming
model-retraining
Tools
python
azure
mlflow
model-deployment
github
Learning Objectives
Develop skills in cleaning, aggregating, and normalizing historical sales, customer preference, and competitive data
Perform Exploratory Data Analysis (EDA) to identify key insights from sales, competitor, and inventory data.
Learn how to derive composite indicators, such as a channel’s overall performance score from raw metrics like units sold, customer preferences, and competitive pressure.
Build and evaluate machine learning models to determine optimal distribution channel.
Grasp the strategic importance of channel selection in launching a new beverage
Integrate MLflow for model tracking, versioning, and reproducibility.
Implement CI/CD pipelines for automating the deployment of machine learning models.
Deploy machine learning models to a production environment using industry-standard tools.
Develop testing strategies for validating machine learning models, including unit tests and model performance evaluations.
Overview
GlobalBev, a prominent player in the beverage industry, is preparing to launch an innovative new beverage line. While the company has access to multiple channels (including e-commerce, Q-commerce, modern trade, general trade, and HoReCa ) launching the product in every channel is not optimal.
In today’s competitive market, selecting the right channel is crucial for ensuring maximum reach and profitability. GlobalBev needs to leverage its historical sales data, customer preferences, and competitive insights to predict which channel is best suited for the new beverage launch.
Your task as a data scientist is to:
- Analyze historical performance data across channels to understand key factors driving success.
- Evaluate customer preferences by region and channel to identify the most receptive markets.
- Assess competitive pressure in each channel to gauge market dynamics.
- Develop a predictive model to determine the optimal channel for launching the new product.
- Addressing these challenges will empower GlobalBev to strategically focus its launch efforts, reduce wasted resources, and maximize the impact of its new beverage line.
Prerequisites
- Building ML Solutions on Azure ML Studio
- Understanding of managing MLlifecycle using MLFlow
- Working knowledge of Inference Endpoints