學門類別
最新個案
- Leadership Imperatives in an AI World
- Vodafone Idea Merger - Unpacking IS Integration Strategies
- V21 Landmarks Pvt. Ltd: Scaling Newer Heights in Real Estate Entrepreneurship
- Snapchat’s Dilemma: Growth or Financial Sustainability
- Did I Just Cross the Line and Harass a Colleague?
- Predicting the Future Impacts of AI: McLuhan’s Tetrad Framework
- Porsche Drive (A) and (B): Student Spreadsheet
- Porsche Drive (B): Vehicle Subscription Strategy
- TNT Assignment: Financial Ratio Code Cracker
- Winsol: An Opportunity For Solar Expansion
London Hydro Inc.: Evaluating Different Electricity Pricing Schemes
內容大綱
London Hydro, Inc. (London Hydro) must forecast the electricity demand of the firm’s clients. The Ontario Energy Board had just announced the results of a pilot program to introduce a new pricing scheme to residential energy consumers in Ontario. This was the origin of tiered pricing, which was based on overall monthly energy usage. In the past, consumers had all been on a time-of-use plan where energy was more expensive during peak hours and cheaper in lower-demand hours. Thus, London Hydro had to anticipate the potential change in client behaviour and predict the effects of the pricing shift.Using data it had gathered on individual household energy consumption, London Hydro hoped that by forecasting which consumers might shift to the new tiered pricing plan it could gain key insights that would help the firm understand what effects the plan might have on its revenues and on its clients’ consumption behaviours.
學習目標
This exercise can be used in undergraduate and graduate-level courses that cover topics in machine learning and predictive analytics, specifically clustering techniques. The teaching note uses R, though Python can also be used following the same steps. This exercise provides students with the opportunity to leverage their knowledge of machine learning techniques (i.e., clustering) and exploratory analysis techniques to draw insights that can be applied to a real-world scenario. After working through the exercise and assignment questions, students will be able to<ul><li>interpret descriptive analytics output, such as data visualizations and summary statistics;</li><li>use R programming language to apply clustering modelling techniques such as k-means clustering; and</li><li>interpret model results to provide recommendations in a real-world scenario.</li></ul>