Mission assurance is the discipline of making sure a facility keeps doing the few things it can never stop doing, whatever else fails. In a refinery that is cooling, containment, pressure relief and shutdown. In a power plant it is fault clearing, heat removal and the ability to island. In a fusion or advanced-fission plant it is the same list with cryogenics, vacuum and tritium boundaries added. Everything else on site is important; these functions are critical, and the difference decides how the plant is engineered.
Start from the consequence, not the asset
Most security programs start from an inventory: list the digital assets, patch them, segment them, monitor them. That work is necessary and it is not mission assurance. Cyber-Informed Engineering (CIE), the framework Idaho National Laboratory developed for the U.S. Department of Energy, starts from the other end. It asks what physical consequence the owner cannot accept, traces every path (digital, mechanical, procedural) that could produce it, and engineers the plant so that consequence is bounded by physics rather than by the state of the software.
The practical output is a critical function register: each function the plant must never lose, ranked by consequence, with the systems it depends on and the single points that could take it down. Protective relaying appears on it. The advisory optimizer does not. That register becomes the design basis for everything downstream.
The single-system test
The test is simple to state and demanding to pass: every critical function must survive the loss of any one system, and the control network counts as a system. If the emergency shutdown depends on a supervisory server staying healthy, the plant fails the test. If relief depends on a controller receiving a setpoint, it fails. Passing usually means the function has an engineered backstop that is not digital: a relief valve sized for blocked-outlet, a relay acting on measured current, a gravity drain, a hard-wired trip on a dedicated logic solver, a manual line-up the operators have actually walked.
What may fail, and what recovers first
Mission assurance is not a demand that nothing fail. It is a decision about what is allowed to fail: optimization, analytics, remote access, historians, anything that can stop without a physical consequence. Those systems are designed to fail to a known state and kept out of the protective path. The remaining question is recovery order. Power, cooling, protection, control, then supervision is the usual sequence, written down, rehearsed, and backed by known-good configuration baselines held offline so a restart is a procedure rather than an investigation.
How this shows up in Porritt Inc. work
Balance-of-plant controls are specified with machine-protection trip paths outside the supervisory layer and one tag namespace from P&ID to operator screen. NORMEX maps each critical function to the OSHA PSM element and the RAGAGEP that governs it, so the compliance record and the mission-assurance register are the same document. Plant Intelligence reads from the control system and never writes to it. The HELIA fusion balance-of-plant program applies the single-system test from the first drawing. The Cyber-Informed Engineering page lays out the six disciplines in order; the critical-infrastructure page covers how this is delivered for national-laboratory and utility owners.
Cyber-Informed Engineering is a framework published by Idaho National Laboratory for the U.S. Department of Energy. Porritt Inc. is an independent company; references describe the framework applied, not endorsement.