HackDavis20

Executive Statement

-We will create a predictive model able to forecast energy needs for each building on UCD campus.

Outcome

-The forecasting tool will help in future planning of energy needs and consumption. We belive this will help optomize budget allocation and reduce waste.

Features:

1. Forecasting - use ML methods to predict future energy costs and usage

2. Weather normalization (optomizing energy use using weather as a factor #not climate)

3. Further investigate relationships between building features and energy usage. Does an unknown feature (ie number of floors, building age)

4. Test for independence between features and costs

Logistics(tenative, if you want to do something else just roger up):

-Chris/Kevin will focus primarily on visualization and python.

-Grant & Jesus will focus on forecasting and deeper relationships between building features

-Dale will focus on alternative energy costs

-John & Alvin will focus on Weather