search

busi732 / Team-2-2024Fall

This is a project repository for Team 2 members in BUSI 732 Quantitative Research, 2024Fall
0 stars 3 forks source link

Weekly Question for the EnergyCo and Professor #22

Open rr-85 opened 5 days ago

rr-85 commented 5 days ago

Objective Establish a routine of posting a weekly question to EnergyCo and the professor to gather insights, clarify project requirements, and ensure alignment with project goals. This will facilitate continuous communication and support the development of a well-informed solution.

rr-85 commented 5 days ago

Weekly Questions

Date 11-12-2024

  1. Definition of Low and High Seasons:
    Could you clarify how we should define "low" and "high" energy demand seasons? Should these labels be based on market data, energy production/consumption data, seasonal patterns, specific monthly thresholds, or a combination of these factors? Additionally, what criteria should we use to generate these labels effectively?

  2. Integration of Marketing Data:
    Given that the marketing data is available only for a specific day each month at precise hourly intervals, what is the recommended approach for integrating this data with our daily and continuous operational data? Are there preferred methods to ensure this data aligns with the other datasets?

  3. Feature Engineering Post-EDA:
    After completing the Exploratory Data Analysis (EDA), what specific guidelines or recommendations should we follow for feature engineering? Are there key features or transformations that you would expect to see, given the project’s focus on predictive maintenance?

  4. List of Required Architecture Decision Records (ADRs):
    Could you provide a list of all required Architecture Decision Records (ADRs) for this project?

@gibby-ci If you can provide answer to those question that would be great. Thanks!

gibby-ci commented 5 days ago

@rr-85

  1. low and high seasons are based off of price (market data) -> think about the business implication. the value to your energy is worth more - so they charge more to get it running.
  2. first pass is to provide a value to model with, second pass is to improve it if there is value to that improvement
  3. needs to come from the constraints and decision variables
  4. no, part of the ADR grade is how well you pick an ADR :)