Project Details

The primary dataset used here is the "pizza_sales.csv" file. Which contain detail of every pizza sold by the store.

Data Cleaning:
In the initial data preparation phase following tasks were performed:

Data Loading and Inspection
Handling missing values
Data cleaning and formatting

Exploratory Data Analysis:
EDA involved exploring sales data to answer the following questions:

What are the peak sales days/months?
Which pizza categories are top sellers?
Which size of pizza is the highest seller?

Data Analysis:
SQL queries were used to analyze the sales.

Based on the analysis, I recommend the following actions for the store:
Focus on the weekend sales, store can give some offers to the customers during these days.
Same way during summer vacation time, the store can do some marketing to increase sales.
The store can add some new categories to attract customers.

Live Interactive Dashboard

Project Details

Data Sources:
There are five different CSV files used. Which contains details of every booking by various platforms, room types, hotels, and dates.

EDA involved exploring sales data to answer the following questions:
Which days of the week show higher booking compared to others?
Which type of rooms have a high frequency of booking?
Which booking platform gives the highest booking?

Data Analysis:
Using Power Query in Power BI some columns were generated to create measures, after that data modeling was performed between the tables.
To generate different KPIs, DAX measures were created in Power BI for ADR-Average Daily rate | RevPAR-Revenue Per Available Room | DSRN-Daily Sellable Room Nights.

Based on the analysis, I recommend the following actions:
Although the occupancy increases during weekends by around 7% compared to weekdays, the ADR has not increased much. So, hotels can focus on price hikes during weekends or can increase available rooms.

Live Interactive Dashboard

Project Details

The dataset contains details of each of the Road accidents that happened in the two years (2021 & 2022) and has more than 300K rows of information.

EDA involved exploring the data to answer the following questions:
What is the road surface condition where most accidents occur?
Which types of vehicles have the highest number of accidents?
What are the weather conditions and light conditions when accidents happen?

For data analysis, DAX measures where created in Power BI, some of them are shown below:
PY casualties = CALCULATE(SUM(Data[Number_of_Casualties]),SAMEPERIODLASTYEAR('Calendar'[Date]))
CY Casualties = TOTALYTD(SUM(Data[Number_of_Casualties]),'Calendar'[Date])

Based on the analysis, I recommend the following action:
Analysis showed that accidents occurred more in urban areas(62%) compared to rural areas(38%), which may be because urban areas have more vehicles. So, the conclusion should not be based on this.

Live Interactive Dashboard

Project Details

EDA involved exploring sales data to answer the following questions:
What is the sales of the store by month?
Which state contributes most to the store's sales?
What is the channel of the order, gender, and age of the customer?

Utilizing Excel's PivotTable feature, the data was transformed into a dynamic pivot table, followed by the generation of diverse charts and visuals tailored to specific requirements.

Based on the analysis, I propose the following strategic actions for the store:
Introduce additional winter wear during the winter season to mitigate the sales decline and attract more customers.
Expand the product range by introducing new men's collections to enhance the appeal for male customers.
Diversify sales channels beyond major platforms like Amazon, Myntra, and Flipkart, which currently contribute 80% of total sales. Explore other smaller channels to broaden the customer base and strengthen market presence.

Project Details

Data Cleaning Tasks:

Handling Missing Values: Removed NaN values from the DataFrame to ensure data completeness.
Row Filtering: Eliminated rows based on specific conditions, enhancing the dataset's relevance.
Column Type Conversion: Applied necessary transformations using to_numeric, to_datetime, and astype methods to ensure accurate data types.

Data Exploration and Business Questions:
- Best Month for Sales
- Top Selling City
- Optimal Advertisement Time
- Frequently Sold Together Products
- Best-Selling Product Analysis

Methods Employed:

- Concatenation
- Column Manipulation
- String Parsing
- Function Application
- GroupBy and Aggregation

Matplotlib Visualization:

- Graphical Representation
- Customization
- Seaborn Integration