Cyclistic Bike-Share Analysis

Importing the Required Python Libraries

The Python environment was prepared by importing the libraries needed for data manipulation, numerical calculations and visualisation.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
pd.set_option('display.max_columns', None)

Loading and Combining the Monthly Datasets

Multiple monthly CSV files were loaded and combined into one complete DataFrame for the analysis period.

import glob

files = sorted(
    glob.glob(
        "/kaggle/input/datasets/ashwinraphel/"
        "cycle-new-data-set/2025*.csv"
    )
)

df_list = []

for file in files:
    temp_df = pd.read_csv(file)
    df_list.append(temp_df)

df = pd.concat(df_list, ignore_index=True)

print(df.shape)

Understanding the Dataset Structure

The dataset was inspected before cleaning to understand its structure, quality and overall characteristics.

df.shape
df.head()
df.info()
df.isnull().sum().sort_values(ascending=False)

Sample Records

Dataset Structure

Missing Values

Data Cleaning and Preparation

The data was cleaned to ensure that comparisons and visualisations were based on valid and consistent records.

df['started_at'] = pd.to_datetime(
    df['started_at']
)

df['ended_at'] = pd.to_datetime(
    df['ended_at']
)

df['ride_length'] = (
    df['ended_at'] - df['started_at']
).dt.total_seconds() / 60

df = df[df['ride_length'] > 0]
df = df[df['ride_length'] <= 1440]

Creating New Analytical Features

New variables were created from the timestamps to support deeper behavioural comparisons.

df['ride_length'] = (
    df['ended_at'] - df['started_at']
).dt.total_seconds() / 60

df['day_of_week'] = (
    df['started_at'].dt.day_name()
)

df['month'] = (
    df['started_at'].dt.month_name()
)

df['hour'] = (
    df['started_at'].dt.hour
)

df['is_weekend'] = (
    df['day_of_week'].isin(
        ['Saturday', 'Sunday']
    )
)

Comparing Member and Casual Rider Behaviour

Rider groups were compared across ride frequency, duration, weekday activity, monthly trends and bike preferences.

Ride Count by Rider Type

Observation

Members completed approximately 3.55 million rides, compared with approximately 2 million casual rides.

Business Insight

Members demonstrate more frequent usage and form Cyclistic's core customer group.

Average Ride Duration

Observation

Casual rides averaged around 19.1 minutes, compared with 12 minutes for members.

Business Insight

Casual riders appear more likely to use bikes for recreation and leisure.

Usage by Weekday

Monthly Ride Trends

Weekend vs Weekday Patterns

Bike-Type Preferences