How is the Global Automation Score (GAS) calculated?

The Global Automation Score anchors to measured IFR World Robotics robot density (robots per 10,000 manufacturing workers, via Our World in Data) and adjusts it by up to ±15% based on 30-day live automation-news momentum from our Signal Engine. Unsourced estimates such as AV fleet size are not score inputs.

What is the Global Automation Pulse (GAP) and how is it calculated?

The Global Automation Pulse is a daily 0-100 index measuring real-world automation signal activity versus a 30-day baseline. Signals are weighted by category: financial events (3×), deployments (2×), leadership changes (1.5×), and workforce events (1×). A score of 50 is an average day; 100 is twice the baseline activity.

Intelligence Methodology

Roboradar24 provides a real-time window into global automation. Here is how we source, calculate, and verify our data.

1. Robot Density Index

Robot density is the number of multipurpose industrial robots per 10,000 employees in manufacturing. Figures are the latest published International Federation of Robotics (IFR) World Robotics data, sourced via Our World in Data and refreshed automatically whenever IFR updates (currently 2024 data). IFR tracks roughly 21 countries; any country IFR does not cover is shown as an estimate and clearly flagged — never presented as a measured figure. Figures follow the IFR/OWID series; where a country's manufacturing-workforce denominator is disputed — most notably China — this can differ from IFR press-release headlines.

2. Global Automation Score (GAS)

The GAS is a proprietary composite index weighted to reflect the full spectrum of autonomy infrastructure:

GAS = RD_IFR × (1 + 0.15 × signal_momentum)

RD_IFR: Robot Density — industrial robots per 10,000 manufacturing workers, IFR World Robotics via Our World in Data (measured; countries IFR does not track use a flagged estimate)
signal_momentum: (30-day scraped automation-signal count ÷ baseline) − 1, from our live Signal Engine — bounds the adjustment to ±15%
Deliberately excluded: AV-fleet and datacenter figures — they have no verifiable per-country source, so they are displayed as clearly-flagged estimates but never move the score

3. Anomaly Detection Engine

We use a Z-score statistical model to detect spikes in automation signals. A country is flagged for "Anomaly" if its latest signal frequency or density shift exceeds 1.8 standard deviations from its 10-period historical average.

4. Data Sources

Our pipeline uses:

Robot density — IFR World Robotics, via Our World in Data; fetched automatically, latest available year.
Live signals & Automation Pulse — real-time scraping of Google News and industry press releases (our Signal Engine).
Supercomputing capacity — TOP500 list, June 2026, fetched automatically from the newest published TOP500 list (updates each June and November). A count of 0 means the country has no system on the current list.
AV fleet — currently a modeled estimate, clearly flagged; migrating to our live deployment tracker and not presented as measured until it is.

Looking for the raw data behind these charts?
Download Latest CSV (Pro Only) →