This Nptel session provides some insights into Automotive Product Development. Link
Use Case #1 - Text Mining - Understand the edge for Top Selling Cars - Collect data, scrap. Find key topics discussed by customers from reviews, feedback, complaints, What features create the buzz for the product
Key Techniques - Sentiment Analysis, Topic Analysis, Features Vs Sales Correlation. Ranking based on customer feedback.
Sales Numbers vs Features - Decision Tree to map sales numbers vs ranking of features, Classification Algorithm based on available data to map and look at feature impact vs Sales
Insights based on Crash Data Analysis
Use Cases #2 - Classification model for Severity of Injury based on different parameters collected
Use Case #3 - Regression model for mileage prediction
Scarping all product-related content from the web is key to understand
- Evaluating new features in the next version
- Comparing with Nearest Competitor
- Evaluating Sentiment post-sales
Additional use cases
- Material level forecasting models based on Service Patterns / Customer Base
- Seasonality based Product segmentation based on failure rates, criticality
- Perform Cluster level forecasting models
- New Product Launch Aspects and related forecasts on sales/service
A lot of use cases, A lot of innovation!!!
Happy Learning!!!
No comments:
Post a Comment