Chris Johnson
Mars Polar Lander /Mars Climate Orbiter resources


In 2001, I conducted a causal analysis of the Mars Surveyor'98 project. The original NASA reports, cited below, provide a thorough analysis of the failures that contributed to the loss of these missions. However, NASA's mishap guidelines NPG 8621.1 were not in force during these investigations. As a result, neither report provides any direct evidence about their use of one of the recommended approaches. The purpose of this study was to see what might have emmerged through the use of change analysis, root cause analysis, MORT and a number of other `advanced' incident analysis techniques.

This project was initially developed after discussions with Michael Holloway at NASA Langley. However, it is independent research and forms part of a forthcoming book on `mishap' analysis in safety-critical systems. My analysis is post hoc and hence is biased by the results of the previous enquiries. However, it does for the first time demonstrate that techniques such as ECF charting, Barrier analysis, Change analysis, Tier analysis and Non-compliance analysis can be used to guide the analysis of mission-critical failures in high-technology systems. It is important to stress that this finding was made in the face of considerable, initial skepticism about the utility of the techniques advocated in NPG 8621.1.

The book chapter/technical report (pdf)


NASA resources

  1. MCO_report.pdf
    NASA Mars Climate Orbiter Report (part I).

  2. MCO_MIB_Report.pdf
    NASA Mars Climate Orbiter Report focussing on project management issues and lessons learnt (parts I and II).

  3. mpl_report.pdf
    NASA Mars Polar Lander and Deep Space 2 Report.

  4. Mpiat_report.pdf
    NASA Mars Program Independent Assessment Team

  5. N_PG_8621_0001_.doc
    NASA procedures and guidelines for mishap reporting, investigating and record keeping.

  6. Administrator Goldin's remarks to JPL following MS'98 Losses

Analytical References

These references describe a subset of the techniques that are recommended by the NASA procedures and guidelines for mishap reporting (NPG 8621.1). As can be seen, these techniques have been developed to support the analysis of workplace injuries rather than the failure of complex, technological systems. A major portion of the work in this project has been to devise ways of mapping the elements of ECF analysis to the particular demands of aerospace applications.

  1. DoE Accident Investigation Handbook (esp Ch7. Data Analysis)

  2. Events and Causal Factors Analysis

  3. Barrier Analysis

Image of Mars Mission Badge
Chris Johnson, Dept. of Computing Science, Univ. of Glasgow, Glasgow, G12 8QQ, Scotland.
Tel: +44 141 330 6053, Fax: +44 141 330 4913, johnson@dcs.gla.ac.uk