Back Portfolio
  • Project Type General Project
  • Category Web Application | Machine Learning
  • Year 2018

The general idea of the project is to design and develop a decision-support system for forecsating campus water demand and management.

In technical, three machine learning algorithms (linear regression, neural network, support vector machine) were adopted to tackle water management problems in different levels, operational, tactical, and strategic problems. Then we visualized the results in a web-based decision-support dashboard.

We used weather-related dataset, water consumption as inputs for the model training process. The key finding of this project is one of the most important factors (from our initual input pool) is The number of raining days, which is related to human behaviors. People in campus tend to consume lots of water after raining for bath and cleaning (our campus has a lot of trees).