The world of data visualization presents a myriad of tools for presenting information in a visually appealing and meaningful way. Among these tools, Sankey diagrams, or Sankey charts as they are sometimes called, are becoming increasingly popular due to their effectiveness in illustrating complex flow patterns. In this article, we will explore how to create Sankey charts and touch upon some of their diverse applications in a journey through Sankey chart adventures.
What Are Sankey Charts?
A Sankey diagram is a type of flow diagram that uses arrows (or “sankey arrows”) to represent the direction and magnitude of flows between nodes. The width of the arrows indicates the capacity or quantity of the flow. These diagrams are named after Captain Matthew Henry Phineas Riall Sankey, a British engineer who used flow diagrams to illustrate the energy transfer in steam engines, which earned him a gold medal from the Institution of Mechanical Engineers.
How to Create Sankey Charts
Creating a Sankey chart involves several steps that vary depending on the software or tool you use (e.g., Excel, Tableau, Python libraries like Plotly or Bokeh). Here’s a general guide on how to create a basic Sankey diagram using Python and the library networkx and pygal.
Step 1: Data Preparation
First, you need to organize your data in columns for the following:
source: Label of the node where the flow starts.target: Label of the node where the flow ends.value: The amount or quantity of the flow.
For instance, you might have the following data:
| source | target | value |
|----------|----------|------------|
| Node A | Node B | 10 |
| Node A | Node C | 20 |
| Node C | Node D | 30 |
Step 2: Setting Up Your Library
Import the necessary libraries in your Python script:
python
import networkx as nx
import matplotlib.pyplot as plt
from pygal_maps_world import reducers as reducers
Step 3: Create and Add Edges to the Network
Use networkx to create a directed graph and add the edges:
python
G = nx.DiGraph()
G.add_edge('Node A', 'Node B', weight=10)
G.add_edge('Node A', 'Node C', weight=20)
G.add_edge('Node C', 'Node D', weight=30)
Step 4: Normalize the Data Before Rendering
Since pygal does not directly support edge weights with arrows, we can convert weight into the percentage of the total flow:
python
values = {source + ':' + target: data['value'] for source, target, data in G.edges(data=True)}
total_flow = sum(values.values())
normalized_values = {k: v / total_flow for k, v in values.items()}
Step 5: Rendering the Sankey Diagram
Use pygal to create the Sankey diagram:
python
pie_chart = pygal.SolidSankey(height=800, width=800, force_log_y=True)
pie_chart.title = 'Flow Diagram'
values = [normalized_values[f'{source}:{target}'] for source, target, data in G.edges(data=True)]
pie_chart.add('Flow', values)
pie_chart.render_to_file('sankey_diagram.svg')
Applications of Sankey Charts
Sankey charts find applications in a variety of fields where flows need to be visualized and understood:
- Environmental Science: Tracking waste streams in recycling, pollution, and energy consumption.
- Economics: Demonstrating supply chains and trade flows.
- Energy Management: Illustrating electrical or heat flows within a building or across districts.
- Sociology: Mapping migration patterns or social interactions within networks.
- Medical Research: Charting molecular or protein pathways through a cell.
- Marketing: Showing customer journey through a sales funnel.
Sankey charts excelled in providing a clear narrative of complex data, making them indispensable in presentations, reports, and user interfaces where comprehensible information presentation is key. As you navigate through the “Sankey chart adventures” in your own projects and applications, remember the journey can lead to innovative insights and deeper understanding of your data flows.
SankeyMaster
SankeyMaster is your go-to tool for creating complex Sankey charts . Easily enter data and create Sankey charts that accurately reveal intricate data relationships.
