Coffee Dashboard

An interactive visualization dashboard for exploring global coffee production, consumption, import, export, and trade flow data from 1990 to 2019.

Overview

This dashboard provides a comprehensive analysis of coffee market dynamics through interactive visualizations built with Dash and Plotly. The dashboard offers detailed insights into production patterns, consumption trends, import/export data, and global trade flows, allowing users to explore coffee market dynamics across different countries and years.

Features

• Production Analysis: Explore coffee production by country, coffee type, and year
• Consumption Analysis: Analyze coffee consumption patterns across countries and time
• Import Analysis: Examine coffee import trends and top importing countries
• Export Analysis: View coffee export data by country and growth rates
• Trade Flow Visualization: Interactive Sankey diagram showing coffee trade flows between countries with filtering by exporter and importer

Visualizations & Technical Details

Common Features Across Tabs

• Year Slider: Interactive slider (1990-2019) using dcc.Slider for temporal filtering
• Coffee Color Theme: Custom color palette with COFFEE_COLORS and COFFEE_COLORSCALE variables

Production Tab

1. Treemap Chart: Country-based production proportional visualization

   • Library: plotly.express.treemap
   • Data Source: Coffee_production_modified.csv
   • Technique: Hierarchical data segmentation by country with dynamic color scaling
   • Features: Interactive tooltips, zoom capability

2. Countries Table: Ranked list of producing countries

   • Implementation: Custom HTML table with html.Table, html.Tr, html.Td components
   • Styling: Alternating row colors, responsive layout
   • Data: Sorted by production volume using sort_values in pandas

3. Radial/Pie Chart: Top 10 producers visualization

   • Library: plotly.express.pie with hole parameter for donut effect
   • Features: "Others" category aggregation for non-top countries
   • Data Processing: get_top_countries helper function to extract top producers

4. Trend Line Chart: Production across years

   • Libraries: plotly.graph_objects.Scatter for line and markers
   • Technique: Polynomial fitting with np.polyfit for trend line
   • Data Processing: Aggregation with get_annual_totals function

5. Coffee Types Bar Chart: Production by coffee variety

   • Library: plotly.graph_objects.Bar
   • Data Processing: get_coffee_type_totals for categorization by Arabica, Robusta, and mixed varieties
   • Features: Value labels with formatted numbers

Consumption Tab

Similar visualizations as Production tab, using Coffee_domestic_consumption_modified.csv with consumption-specific calculations and titles.

Import Tab

1. Treemap Chart: Country-based import proportional visualization

   • Data Source: Coffee_import.csv
   • Implementation: Same technique as production treemap with import data

2. Countries Table: Ranked list of importing countries

   • Sorted by import volume using pandas

3. Radial Chart: Top 10 importers visualization

   • Same implementation as production but with import data

4. Trend Line Chart: Import volumes across years

   • Shows historical import patterns with trend line

Export Tab

1. Treemap Chart: Country-based export proportional visualization

   • Data Source: Coffee_export.csv with special handling for missing/invalid values
   • Data Cleaning: replace(-2147483648, np.nan) to handle placeholder values

2. Countries Table: Ranked list of exporting countries

   • Implementation: Similar to import/production with export-specific data

3. Radial Chart: Top 10 exporters visualization

   • Data Processing: Includes filtering for NaN values with dropna()

4. Trend Line Chart: Export volumes across years

   • Features: Special handling for missing data in trend calculation with fillna(0)

Trade Flow Tab

1. Filter Dropdowns: Country-specific filtering

   • Components: dcc.Dropdown for exporter and importer selection
   • Implementation: Dynamic filtering based on selection

2. Sankey Diagram: Flow visualization of trade connections

   • Library: plotly.graph_objects.Sankey
   • Data Source: synthetic_coffee_trade_flows.csv
   • Features: Node coloring by type (exporter/importer), dynamic sizing based on quantity

3. Top Countries Tables: Exporters and importers rankings

   • Implementation: Dynamic table generation based on filtered dataset

Data Sources & Processing

The dashboard uses the following CSV data files:

• Coffee_production_modified.csv: Production volumes by country and coffee type
• Coffee_domestic_consumption_modified.csv: Consumption data by country
• Coffee_import.csv: Import volumes by country
• Coffee_export.csv: Export volumes by country (with special handling for -2147483648 values)
• synthetic_coffee_trade_flows.csv: Trade flow connections between countries

Data processing techniques used:

• Pandas operations: groupby, sort_values, sum, dropna, replace, merge
• NumPy functions: polyfit, poly1d, isnan, percentile
• Custom helper functions:
  • get_top_countries: Extracts top N countries for a specified metric and year
  • get_annual_totals: Calculates yearly totals with optional coffee type filtering
  • get_coffee_type_totals: Segments data by coffee type (Arabica, Robusta, mixed)
  • get_production_consumption_by_year: Creates comparative yearly data

Technologies Used

• Dash: Web application framework for creating interactive dashboards
• Plotly: Interactive visualization library
• Pandas: Data manipulation and analysis
• NumPy: Numerical computations and array operations
• Python 3: Core programming language

Setup and Running

sed -i '' 's/\r$//' ./run_final_dashboard_fixes.sh

Running the Dashboard

./run_final_dashboard_fixes.sh

Or directly with Python:

python coffee_dashboard_revised.py

Environment Setup

Method 1: Using setup_env.sh (Recommended)

./setup_env.sh

Method 2: Using pip

pip install -r requirements.txt

Method 3: Using conda with environment.yml

conda env create -f environment.yml
conda activate coffee_dashboard

After starting the dashboard, open your browser and navigate to:

http://127.0.0.1:8050/

Browser & Interface Notes

• For best performance, use Chrome, Firefox, or Edge
• The dashboard is responsive and works on both desktop and mobile devices
• All charts include interactive features:
  • Hover tooltips with detailed data
  • Zoom/pan capabilities
  • Click interactions for filtering
  • Download as PNG option

Key Files

• coffee_dashboard_revised.py: Main dashboard application
• run_final_dashboard_fixes.sh: Dashboard launch script
• CSV data files: Source data for visualizations
• requirements.txt and environment.yml: Dependency specifications