The negative impacts of flight delays and disruptions are well documented. Globally, flight delays...

# What do the Lumo Delay Indexes mean?

The Lumo delay indexes are a score from 1 to 10 given to each flight indicating how “risky” a flight is with respect to being delayed. The score is intended to capture both delay frequency (probability of a delay occurring) and delay severity (how long will the delay be if it does happen).

Quite simply, a delay index of 1 indicates near-certainty of less than a 30-minute delay, while an index of 10 indicates near-certain delay of 2 hours or more; the numbers in between are essentially a weighted average of the likelihood of a delay of different magnitudes.

Lumo calculates this index as a weighed average of 4 numbers:

- The likelihood of a delay of less than 30 minutes (p0)
- The likelihood of a 30-60 minute delay (p1)
- The likelihood of a 1-2 hr delay (p2)
- The likelihood of a 2+ hour delay (p3)

A delay index of 1 corresponds to the case when the first bucket (p0) is close to 100%. A delay index of 10 corresponds to the case when p3 is close to 100%. A certain delay of at least 30 minutes results in a delay index of 6.

Here’s a breakdown of the delay index 16 hours out vs the average delay for 1 year of flight data. Obviously, these are averages and your specific situation might be different.

Here's one example of the predictions.

At 12 hours before departure, Lumo alerts the traveler that winds at LGA could disrupt their flight, and assigns a delay index of 5 (there is a 60% chance of less than a 30-minute delay at that point).

Subsequently, as conditions evolve, Lumo upgrades the prediction to a 7 (only a 25% chance of less than a 30-minute delay).

The flight eventually ends up being significantly delayed (since there is a 100% chance of a 2+ hour delay at this point, the delay index would be 10).

Here is another help page on the delay indexes and a blog post on the factors that go into our predictions.