Bad Data: How Inaccuracies Can Kill Your Mission

By Jorge Ruiz on May 25, 2016

This image is the first view of Mars taken by the Mars Climate Orbiter on September 7th, 1999, when the spacecraft was approximately 4.5 million kilometers (2.8 million miles) from the planet.Mars from the Mars Climate Orbiter

Unfortunately this was the first and only image taken by the Mars Climate Orbiter. After investing a total of $327.6 million in this mission, including $193.1 million for spacecraft development, $91.7 million for launching it, and $42.8 million for mission operations, a simple metric mismatch caused the mission to fail. Due to complications stemmed from human error, the NASA spacecraft encountered Mars at an altitude of 57 kilometers of the surface instead of 226 kilometers and disintegrated due to atmospheric stresses.

The primary cause noted in the investigation was that software supplied by Lockheed Martin used measurements in a United States customary unit, when NASA’s system at their Jet Propulsion Laboratory in Pasadena expected those results to be in metric units, in accordance with the Software Interface Specifications (SIS). The software that calculated the total impulse produced by thruster firings calculated results in pound-seconds. The trajectory calculation software then used these results – expected to be in newton-seconds – to update the predicted position of the spacecraft.

Lockheed Martin helped build, develop and operate the spacecraft for NASA – should they have realized those errors in the reporting? Maybe, but NASA didn’t point any fingers. “This is an end-to-end process problem,” said Tom Gavin, the person to whom all project managers at the Jet Propulsion Lab report. “A single error like this should not have caused the loss of Climate Orbiter. Something went wrong in our system processes in checks and balances that we should have caught and fixed.”

How was the data transferred? How did it originally get into system in English units? With a whole team doing navigation,distance and speed checks, how is it possible that they missed it?

Everyone is susceptible to human errors. But, as Gavin said, the problem here was not the error, “It was the failure of us to look at it end-to-end and find it.”  Mistakes even occur on multi-million spacecraft projects – however sufficient processes should be in place to identify errors before they become critical to mission success. If this can happen to NASA, with a 19 billion dollar budget over several years of planning by some of the brightest minds in the world – do you think this can happen to your company? Do you verify the consistent use of units throughout your different systems?

Do you conduct audits for specification compliance on all data transfers? Do you trust your verification and validation processes?

Questions or comments about this article?

Tweet @Winshuttle to continue the conversation!