In the 2006 timeframe NASA’s Constellation Program was looking to address several issues with the way Problem Reporting data had been collected for the Shuttle Program including multiple systems across groups and centers (20+ for Shuttle), inconsistent schemas and processes across systems, difficulty searching within each system, and lack of ability to search across systems. The Program’s goal was to deploy a single new system to be used across the participating groups and centers. The Ames Human-Computer Interaction (HCI) group conducted requirements research into problem reporting across NASA missions (Shuttle, ISS, Mars Exploration Rovers, etc.), centers (JSC, KSC, MSFC, Langley, Armstrong, etc.), and external groups (e.g., the Navy’s SubSafe Program).

The requirements research showed that problem reports, on-orbit anomalies, and similar complex data sets had historically been stored in software, either custom built or off-the-shelf, with a fixed data schema (a set of fields and values) and process. However, mission needs and processes changed over time, differed between individual missions, and needed to support specific local needs within larger, more complex missions. Changes to fixed data schemas required software development. Missions had to invest money and then wait for developers to implement changes in order to adapt the tools to their processes. Without software development, the tools defined and constrained mission processes. A new system for Constellation would need to support both local variations in process as well as evolve over the mission lifecycle without expensive and time-consuming changes to code.

By 2006, building software with a configurable schema and process (workflow) had become less expensive: 1) open source components were available to limit the amount of software development required, 2) the Web could support complex applications for data entry and search from any location, and 3) hardware improvements enabled fast enough browser and server performance to run more complex, configurable software. In 2007, the HCI group deployed Constellation PRACA (Problem Reporting Analysis and Corrective Action), the first instance of Mission Assurance Systems (MAS), which met the Program’s goals. Constellation PRACA was adopted by KSC, Glenn, Langley, MSFC, and other groups. Shortly after the successful deployment of CxPRACA, the Constellation Program asked the HCI group to evaluate whether the MAS software, with its configurable schema and process, could be used for other data sets: Hazards, Failure Modes Effects Analyses, etc. Following additional requirements research, the HCI group deployed three additional systems for Constellation.

In the next few years, the International Space Station Program, Ames, Human Research Program, White Sands Test Facility, and others deployed additional instances of MAS for a variety of engineering data sets. After the Constellation Program was cancelled in 2011, NASA’s new Human Space Flight Program, Exploration Systems (Space Launch System, Multi- Purpose Crew Vehicle, and Grounds Systems Development and Operations), adopted the Constellation systems and deployed a number of new systems and integrations between those systems. Currently there are 19 MAS production systems and many integrations between those and external systems. MAS is now used at an Agency level with improvements to data management in the areas of process efficiency (reduced costs) and data accuracy.

This work was done by Tina Panontin, Alonso Vera, Irene Tollinger, Christian Ratterman, Matt Sharpe, Alex Eiser, Ken Ebbs, Guy Pyrzak, and Collin Green of Ames Research Center; and Matt Guibert, Kristle McCracken, Kevin McMillin, Joe Medwid, Chris Mueller, Stephen Trahan, Dave Luetger, Jill Foley, Matt Johnson, Gary Boal, Alan McCreary, Maria Suarez, Anna Jacinto, and Jay Grewal of San Jose State University. This software is available for use. To request a copy, please visit https://software.nasa.gov/software/ARC-16033-1 


Topics:
Software