[bar-tornado-sensitivity] Tornado Diagram for Sensitivity Analysis

[MarkusNeusinger]

Description

A horizontal bar chart with bars sorted by influence magnitude, extending left and right from a baseline value. Each bar shows how varying one input parameter affects the output, creating a tornado-like shape (widest at top, narrowest at bottom).

Key Visual Elements

• Horizontal bars centered on base case value
• Bars sorted by total range (widest at top)
• Left side: low input value effect; right side: high input value effect
• Base case vertical reference line
• Parameter names on y-axis
• Dual colors for low/high scenarios

Applications

• Financial modeling: identifying key value drivers
• Risk analysis: one-way sensitivity analysis
• Project management: assessing cost/schedule uncertainty drivers
• Engineering: identifying critical design parameters
• Consulting: communicating sensitivity results to stakeholders

Domain

Risk Analysis / Decision Science

[claude]

✅ Spec ID: bar-tornado-sensitivity

Rationale: This is a valid and widely-used visualization in risk analysis and decision science. While bar-diverging exists, it covers general positive/negative divergence from zero — a tornado diagram is a distinct variant specifically designed for sensitivity analysis, with bars sorted by influence magnitude and dual-colored low/high parameter effects extending from a base case reference line. No duplicate found.

Add label approved to begin generation.

[github-actions]

✅ Specification Ready: bar-tornado-sensitivity

PR: #4595

---

specification.md

bar-tornado-sensitivity: Tornado Diagram for Sensitivity Analysis

Description

A horizontal bar chart where bars are sorted by influence magnitude, extending left and right from a base case vertical reference line. Each bar represents one input parameter and shows how varying that parameter between its low and high values affects the output, creating a characteristic tornado shape (widest bars at top, narrowest at bottom). Dual colors distinguish low-input from high-input effects, making it immediately clear which parameters drive the most uncertainty.

Applications

• Financial modeling: identifying which assumptions (discount rate, growth rate, costs) most affect net present value
• Risk analysis: performing one-way sensitivity analysis to rank uncertainty drivers in project cost or schedule estimates
• Engineering design: determining which design parameters have the greatest impact on system performance
• Consulting and stakeholder communication: presenting sensitivity results in an intuitive, ranked format

Data

• parameter (str) - Name of the input variable being varied (e.g., "Discount Rate", "Material Cost")
• low_value (float) - Output result when the parameter is set to its low scenario value
• high_value (float) - Output result when the parameter is set to its high scenario value
• base_value (float) - Output result at the base case (single value used as the center reference line)
• Size: 6-15 parameters recommended for readability; bars sorted by total range (high - low) descending
• Example: NPV sensitivity analysis with 8-10 financial assumptions varied one at a time

Notes

• Draw a vertical reference line at the base case value
• Sort bars by total range (|high_value - low_value|) with the widest bar at the top
• Use two distinct colors: one for the low-scenario side and one for the high-scenario side
• Label each bar with the parameter name on the y-axis
• Optionally display the low/high input values or resulting output values at bar ends
• The x-axis represents the output metric (e.g., NPV, cost, duration)

---

Next: Add approved label to merge this specification to main.

---

🤖 spec-create workflow

[github-actions]

🚀 Specification Merged!

Specification: bar-tornado-sensitivity
PR: #4595 (merged)

The specification is now in main and synced to PostgreSQL.

---

Next: Add generate:{library} labels to start generating implementations.

Available libraries: matplotlib, seaborn, plotly, bokeh, altair, plotnine, pygal, highcharts, letsplot

Or use workflow dispatch for bulk operations.

---

🤖 spec-create workflow