Project Motivation

Energy Consumption in Singapore

To align with the government's goal to reduce energy consumption and wastage and improve public awareness and education on issue, our team sets out to explore ways to improve the way energy consumption data is being visualized and displayed.

Issues/Problems

Lack of coherent visualization

While energy consumption data in Singapore is readily available, the visualization lacks interactivity and coherence, resulting in users being unable to gain useful insights from the visualization.

Approaches

Data Preparation and Visualization

The datasets are stacked, concatenated and geocoded. To visualize the dataset, we made use of D3.js to present the data using a variety of visualization techniques.

Electrical Consumption by Housing Type in 2015: Line Chart

Based on the line chart above, the following can be observed:

1. The bigger the housing type, the higher the electrical consumption. E.g a 5-room/executive type will consume more electricity then that of a 3-room type.

2. Peak electrical consumption seems to be the highest in the month of August across all housing types.

3. Electrical consumption is at its lowest in months of February and March.

Consumption Per Room By Housing Type: Line Chart

While it was previously observed that the more rooms there are, the higher the energy consumption, we wondered if the same observation applies when we based the comparison on the average room consumption for each housing type.

By calculating the average energy consumption of each room in a house across all housing types, it was observed that 4-room public housing tend to have a higher energy consumption, and it seems that it is consistently higher across the year as compared to the bigger 5-room/executive housing.

Comparison of Housing Type Units against Annual Consumption: Double Bar Chart

A double bar chart was used to compare the composition of the housing types in 2015 against its annual consumption. Over here we are able to observe that the majority of housing in 2015 belonged to 4-room and 5-room / Executive types.

While we observed that the the most common type of public housing in Singapore would be the 4-room type, the 5-room and executive housing actually have a higher total annual electricity consumption.

It can also be observed that while 1-room / 2-room flats are the smallest group of houses, the electricity consumption is disproportionate to their size.

Subzone/Region Analysis: Bubble Graph

To further analyse the electricity consumption in Singapore, the group decided to make use of regions as well as subzones to observe if there are any particular regions or subzones with high electrical consumption.

It can be observed that Boon Lay has the highest electrical consumption amongst all the subzones, with Changi Village being the least.

By rearranging the bubbles into their respective regions, we can also see that the West Coast region has the least electrical consumption amongst all the regions in Singapore, and that the West and East Regions use up more electricity than the rest.

Planning Area Analysis: Treemap

The treemap provides a hierarchical view of the electrical consumption across the various regions, planning areas and subzones in Singapore.

At one glance, it is easy to tell that the West area has the highest consumption and that the Jurong planning area accounts for approximately half of the energy consumption of the West area.

In the event that it is difficult to tell by size the electrical consumption comparison, the colour intensity will help the user to see the differences and extremities in terms of the electricity usage across all the zones.

Demographics Analysis: Treemap

Going further, we attempted to see if demographics (age group) plays a part in the energy consumption, where we had a conjecture that the higher the number of working population in an area, the more the energy consumed. Once again a tree map is used, and the size of the tree map represents the energy consumption while the colour intensity represents the percentage of active working people.

It seems that going by region and the percentage of active working people, we can see a relationship where the region with the highest consumption has the higher proportion of the active working population, and west coast (consuming least energy) has the least proportion of active working people.

However, once we go deeper into the various planning area and subzones, there does not seem to be much of a correlation between the colour intensity and size of the tree map. Hence, there should instead be other factors influencing the energy consumption in Singapore.

Team KABAK

Project Manager

School of Business

Coder

School of Information Systems

Coder

School of Information Systems