1. Abstract

Air pollution has become one of the major environmental and health concerns in many metropolitan cities around the world. Monitoring and analysing air quality data helps researchers and policymakers understand pollution patterns and take necessary measures to improve environmental conditions. Beijing, being a highly populated and industrialized city, experiences significant air pollution levels, making it an important case study for air quality analysis.

This project focuses on analysing the Beijing Air Quality dataset obtained from the UCI Machine Learning Repository. The dataset contains hourly measurements of several atmospheric pollutants such as carbon monoxide (CO), benzene, nitrogen dioxide, and other gases, along with environmental factors like temperature, relative humidity, and absolute humidity.

Using Power BI, the dataset is first cleaned and transformed using the Power Query Editor. Data preprocessing steps such as handling missing values, correcting date and time formats, renaming columns, and removing unnecessary columns are performed to prepare the dataset for analysis.

After preprocessing, interactive visualizations are created in Power BI to analyse pollution trends, relationships between different pollutants, and the impact of environmental factors such as temperature and humidity on air quality. This project demonstrates how business intelligence tools like Power BI can be effectively used for environmental data analysis and visualization.

2. Objectives

The main objectives of this project are:

1. To understand the importance of air quality monitoring and environmental data analysis.
2. To study the Beijing air quality dataset obtained from the UCI Machine Learning Repository.
3. To clean and preprocess raw environmental data using Power Query Editor.
4. To handle missing values and incorrect data entries in the dataset.
5. To transform and organize time-series data for analysis.
6. To create meaningful and interactive visualizations using Power BI.
7. To analyse relationships between pollutants and environmental factors such as temperature and humidity.
8. To develop an informative dashboard for air quality monitoring and analysis.

 3. Existing System

Traditional air quality monitoring systems mainly rely on:

1. Manual environmental data collection and reporting
2. Basic spreadsheet analysis tools
3. Static reports for pollution monitoring
4. Limited visualization and interpretation of environmental data

Limitations of Existing Systems

1. Difficulty in analysing large environmental datasets efficiently.
2. Limited data visualization capabilities.
3. Time-consuming manual data cleaning processes.
4. Static reports that do not support interactive analysis.
5. Limited ability to identify patterns and correlations in pollution data.

These limitations highlight the need for advanced data visualization and analytics tools such as Power BI.

4. Proposed System

The proposed system uses Power BI to analyse and visualize the Beijing air quality dataset.

In this system:

1. The air quality dataset is imported into Power BI.
2. Data transformation and cleaning are performed using Power Query Editor.
3. Issues such as incorrect date and time formats are corrected.
4. Missing values represented by -200 are identified and handled appropriately.
5. Column names are renamed for better understanding.
6. Unnecessary columns are removed from the dataset.
7. The cleaned dataset is used to create interactive dashboards and visualizations.

The system enables users to explore pollution trends, relationships between pollutants, and environmental conditions using interactive Power BI dashboards.

1. Implementation Procedure

The implementation of this project involves the following steps:

Step 1: Data Collection

The dataset is obtained from the UCI Machine Learning Repository. It contains hourly air quality measurements recorded in Beijing between 2004 and 2005.

Step 2: Data Loading

The dataset is imported into Power BI Desktop.

The Transform Data option is used to open the Power Query Editor for data preprocessing.

Step 3: Data Understanding

The dataset includes the following types of information:

1. Date and time information
2. Air pollutant concentrations such as Carbon Monoxide (CO) and Benzene
3. Environmental parameters such as temperature, relative humidity, and absolute humidity

The dataset is identified as time-series data, where measurements are recorded hourly.

 Step 4: Data Cleaning

Several preprocessing steps are performed:

1. Removing incorrect date values in the time column
2. Merging date and time columns for proper time-series representation
3. Handling missing values represented by -200
4. Replacing missing values with appropriate values such as mean or null
5. Removing empty or unnecessary columns (such as column 16 and column 17)

 Step 5: Column Transformation

The columns are renamed to make them more understandable.

Examples include:

1. CO – Carbon Monoxide concentration
2. NMHC – Non-Methane Hydrocarbons
3. C6H6 – Benzene concentration
4. T – Temperature in degrees Celsius
5. RH – Relative Humidity
6. AH – Absolute Humidity

Step 6: Data Visualization

After cleaning the dataset, various visualizations are created using Power BI such as:

1. Line charts for pollution trends over time
2. Bar charts for pollutant comparisons
3. Scatter plots to analyse relationships between temperature and pollution levels
4. Dashboard panels to monitor air quality indicators

 Step 7: Dashboard Creation

An interactive Power BI dashboard is created to present insights from the dataset and enable easy analysis of air pollution patterns.

 6. Software Requirements

The software tools used in this project include:

1. Power BI Desktop – Data visualization and dashboard creation
2. Power Query Editor – Data cleaning and transformation
3. Microsoft Excel / CSV files – Dataset format
4. Windows Operating System

7. Hardware Requirements

Minimum Hardware Requirements:

1. Processor: Intel i3 or higher
2. RAM: 4 GB or higher (8 GB recommended)
3. Storage: 256 GB or higher
4. Laptop or Desktop Computer

8. Advantages of the Project

1. Enables effective analysis of large environmental datasets.
2. Provides interactive dashboards for air quality monitoring.
3. Helps identify pollution trends and environmental patterns.
4. Improves data understanding through visual representation.
5. Supports data-driven decision making for environmental policies.
6. Simplifies complex environmental data using visual analytics.
7. Demonstrates the practical application of Power BI in environmental data analysis.