SAM.
System–Action–Management Framework
Originators Susan M. Murphy & M. Elisabeth Paté-Cornell (Stanford, 1996)
Paradigm PRA extension to organisational factors
Unit of analysis Management → human decisions/actions → system response
Primary domains Aerospace, oil & gas, chemical, maritime, defence

SAM extends classical probabilistic risk analysis (PRA) by linking management factors — incentives, training, procedures, culture, selection — to the decisions and actions of operators, and in turn to the probabilities of technical events in the physical system. It provides a structured way to quantify "why" behind human error in PRA.

Overview of the framework

Murphy & Paté-Cornell (1996) distinguish three coupled layers. The System layer is the conventional PRA model of the physical plant (fault/event trees, reliability data). The Action layer models the decisions and actions of operators, maintenance and management; intention formation is represented by alternative decision models — rational, bounded-rational and rule-based — and execution is modelled separately, allowing errors of intention, execution or both. The Management layer captures organisational policies and incentives that shape which actions people choose to take (e.g., reward structures, staffing, procedures, training). The probability of specified actions is conditioned on management factors; action probabilities then propagate into PRA basic events. Paté-Cornell & Murphy (1996) and Paté-Cornell (2002) illustrate SAM with tanker safety, offshore platforms and space missions.

Management (M) — policies, incentives, training, culture, staffing Shapes which decisions and actions are likely to be taken Human Decisions & Actions (A) Intention: rational | bounded-rational | rule-based · Execution: success / error P(action | M) → feeds basic events in PRA Physical System (S) — fault trees, event trees, component reliability Outcome frequencies & consequence distributions feedback / learning
Figure 1. SAM's three-layer causal architecture: management factors shape human decisions and actions, whose probabilities feed into a classical PRA of the physical system.

When to use it

Typical applications

  • PRA of complex socio-technical systems where organisational factors are material.
  • Evaluation of management interventions (incentives, staffing, training) on risk.
  • Retrospective analysis of accidents with dominant organisational contributors.
  • Teaching framework for organisational reliability and decision analysis.

Aviation & cross-domain relevance

  • Applied to space shuttle tile risk (Paté-Cornell & Fischbeck), offshore platforms, anaesthesia and maritime tanker safety.
  • Influential on modern aviation safety risk models that quantify organisational influence on human error.
  • Conceptual cousin of I-RISK and the Engineering Risk Analysis Method (ERAM).

Benefits

Analytical strengths

  • Provides an explicit, probabilistic mechanism to carry organisational factors into PRA.
  • Accommodates multiple decision models — rational, bounded-rational, rule-based — reducing over-simplification of human behaviour.
  • Separates intention and execution errors, aligning with human-reliability practice.
  • Supports comparative analysis of management interventions.

Practical strengths

  • Theoretically grounded and taught within Stanford's engineering-risk-analysis programme.
  • Flexible: can be layered on top of existing event/fault trees.
  • Useful to communicate why a management action changes risk, not only whether.
  • Well documented case studies across transport, energy and aerospace.

Limitations

  • Data-hungry. Credible probabilities for decision models and organisational influences are scarce.
  • Specification cost. Building three coupled layers is intensive compared to purely technical PRA.
  • Subjectivity. Mapping from management factors to action probabilities relies heavily on expert judgement.
  • Mostly research-based. SAM is widely cited but less frequently implemented end-to-end in industrial practice.
In short SAM is a seminal attempt to put organisational behaviour on a probabilistic footing. It sits alongside I-RISK as one of the canonical methods for linking management quality to quantitative risk.

References (APA 7)

Murphy, D. M., & Paté-Cornell, M. E. (1996). The SAM framework: Modeling the effects of management factors on human behavior in risk analysis. Risk Analysis, 16(4), 501–515. https://doi.org/10.1111/j.1539-6924.1996.tb01096.x

Paté-Cornell, M. E., & Murphy, D. M. (1996). Human and management factors in probabilistic risk analysis: The SAM approach and observations from recent applications. Reliability Engineering & System Safety, 53(2), 115–126. https://doi.org/10.1016/0951-8320(96)00040-3

Paté-Cornell, M. E. (2002). Finding and fixing systems weaknesses: Probabilistic methods and applications of engineering risk analysis. Risk Analysis, 22(2), 319–334.

Davoudian, K., Wu, J.-S., & Apostolakis, G. (1994). Incorporating organizational factors into risk assessment through the analysis of work processes. Reliability Engineering & System Safety, 45(1–2), 85–105. https://doi.org/10.1016/0951-8320(94)90079-5

Paté-Cornell, M. E., & Dillon, R. L. (2001). Probabilistic risk analysis for the NASA space shuttle: A brief history and current work. Reliability Engineering & System Safety, 74(3), 345–352. https://doi.org/10.1016/S0951-8320(01)00063-4

Further reading

Paté-Cornell, M. E. (1990). Organizational aspects of engineering system safety: The case of offshore platforms. Science, 250(4985), 1210–1217. https://doi.org/10.1126/science.250.4985.1210

Paté-Cornell, M. E. (2013). Engineering risk analysis method and some applications. In Advances in risk analysis (Ch. 16). Society for Risk Analysis.

Paté-Cornell, M. E. (2009). Probabilistic risk analysis versus decision analysis: Similarities, differences and illustrations. In Advances in decision analysis. Cambridge University Press.

Øien, K. (2001). A framework for the establishment of organisational risk indicators. Reliability Engineering & System Safety, 74(2), 147–167.

Mohaghegh, Z., Kazemi, R., & Mosleh, A. (2009). Incorporating organizational factors into probabilistic risk assessment (PRA) of complex socio-technical systems: A hybrid technique formalization. Reliability Engineering & System Safety, 94(5), 1000–1018. https://doi.org/10.1016/j.ress.2008.11.006