Project title

Fuel Consumption (TBD)

Project statement

Context

Canadians have built a society that is the envy of the world, but in the process, we've developed a serious problem. Canada uses more energy per capita than almost any other country in the world. One of the reasons for this is our reliance on the automobile. Canadians own about 19 million light-duty vehicles including cars, vans and light-duty trucks, and typically drive more than 300 billion kilometers (km) per year. With close to one vehicle for every two Canadians, we have one of the highest ratios of car ownership in the world.

To a degree, our energy consumption in the transportation sector can be explained by our climate, the vast size of the country and the locations of our population. A great deal of fuel is also wasted in Canada. This happens when we make uninformed purchasing decisions, practice inefficient driving behaviors and fail to properly maintain our vehicles. It's not just a waste of energy – it's a huge waste of money, too. By some estimates, Canadian motorists could save hundreds of dollars per year in fuel and maintenance costs by adopting fuel-efficient practices.

Therefore, gaining a comprehensive understanding of model-specific fuel consumption ratings and the estimated carbon dioxide emissions for new light-duty vehicles intended for retail sale in Canada becomes vitally significant.

Problem Statement

• Exploratory Data Analysis (EDA): Explore the dataset to understand the distribution of fuel consumption ratings across different vehicle attributes. Identify trends and patterns that could affect fuel efficiency.

• Segmentation Analysis: Segment the dataset based on vehicle types, manufacturers, or other relevant factors. Compare fuel consumption trends among different segments and provide recommendations tailored to each segment.

• Visualization: Create compelling visualizations to communicate your findings effectively to a non-technical audience. Visualizations could include fuel efficiency trends over time, comparison between different vehicle types, and more.

Have you noticed any trends or unusual observation about fuel consumption, do mentioned in the slides

Scope of solution space

The sections above were provided to me by Vancouver Datajam, and I was asked to fill in this and the following sections 🙂 Looking forward to working with yall and having fun, no pressure! I want to stress, this is a very beginner friendly and no stress team, if you’re not able to participate and/or contribute and/or produce a deliverable, then don’t worry about it haha.

We are focused on creating an extremely beginner friendly, low code no code environment, where the focus is on fun and exploration rather than being hardcore . If you can manage your way around excel, then great. There are quite a few free all-in-one data cleaning, analyzing, and visualization tools available out there; my favorite, go-to, insanely-beginner-friendly, easy-to-learn tool is TABLEAU. Tableau has a very powerful and free desktop version available to be downloaded from here.

I want to let you all shine without being very techno-deterministic … hence, if you don’t want to use tableau, or excel, just use any tool that you see fit, and/or that you want to experiment with. I want to emphasize that I just want us to simply focus on generating cool/good/great/amazing insights, which we can use to put into our presentation; I don’t want us to be hung up on using xyz tool (software, programming language, etc.).

I rarely say ‘no’, and I like to be fairly informal and dynamic; I’m a big proponent of including most/all work that people do in some way/shape/form into our final deliverable without stifling creativity, for the most part haha. I have given a very general/vague step-by-step process of how we can structure our efforts, but by no means do we have to adhere to it.

1) Data Engineering

1.1) Exploratory Data Analysis
1.2) Data Dictionary ?
1.3) Clean the datasets
1.4) Combine the datasets

2) Data Analysis

2.1) Exploring and Analyzing the data (looking for trends, making new fields/metrics, etc.)
2.2) Insight Generation

3) Presentation (Finalizing our Deliverables)

• Could be us just showing off our insights
  • Trends
  • Patterns
• Could be us showing off our prediction abilities?
  • Maybe we built a prediction model/engine based on
    • type of driver
      • Distance traveled
        • City and/or Highway
        • lifestyle
      • aggressive or easy
• Combination of both and many other things

You can see one example of me having already produced one deliverable after going through the above steps/process/structure by clicking here

Data Dictionary

Data is unstructured-ish and very expansive … this is the fun of this project. We get to learn and explore !

Data Dictionary (preliminary) :

https://www.fueleconomy.gov/feg/ws/

Data

https://www.fueleconomy.gov/feg/download.shtml

• Data from 1995 to 2024 on fuel consumption and other various metrics regarding fuel economy

https://www.kaggle.com/datasets/mohamedhanyyy/canada-cars-sales-figures-20192021

• Canadian car sales data

Project team members

Team lead: Ahmed

Mentor:

Documentation:

Scripting:

Modelling:

Sanity checking:

Tableau Workbook:

Fuel Consumption

Vancouver Datajam 2023 Schedule:

Main page: https://vancouverdatajam.ca/