The questions below are the organizers' initial set. They are candidates, not a program. Prospective attendees are invited to propose additional questions, to argue that one of these should be dropped or merged, or to bring a concrete acceptance decision that tests one of them. Questions marked as priority discussions receive the most time on the one-day program; the remainder shape invited case studies and the closing panel.
How much conservatism accumulates between the environment definition and the acceptance criterion, and which of it covers an identifiable uncertainty.
Follow one requirement through environment selection, shielding assumptions, part variability, test conditions, and acceptance criteria. Where is the same uncertainty covered more than once? Which requirements originate in standards, which in project policy, and which in inherited practice?
Priority discussion: Accumulated margin
Link to this topicWhen should mission duration and an accepted exceedance probability replace a prescribed worst-case event? Cumulative fluence, maximum-event fluence, and peak flux answer different questions and should be treated separately. How should an environmental exceedance probability translate into an acceptable mission risk?
Priority discussion: Statistical environments
Link to this topicWhen are simple shielding geometries adequate, and when does detailed sector analysis or particle transport change the decision? Model fidelity, material assumptions, local shielding, and validation all bear on the answer. What uncertainty remains after a detailed calculation, and how should that uncertainty affect margin?
Priority discussion: Transport and radiation design margin
Link to this topicCan radiation design margin be selected from the uncertainties in the environment, the transport calculation, and the device response, rather than applied as a fixed factor? When is a fixed factor defensible? Published statistical radiation hardness assurance methods provide a foundation for combining environment distributions with part variability to estimate reliability at a stated confidence.
Priority discussion: Transport and radiation design margin
Link to this topicThis includes gamma testing and the dose accumulated during proton SEE testing. What dosimetry, parameter measurements, bias conditions, and dose-rate characterization make those results usable for acceptance? For enhanced low-dose-rate sensitivity, what evidence would support a device-specific bounding factor or an accelerated method? A generic ELDRS multiplier is itself a claim that requires validation.
Priority discussion: High-dose-rate TID credit
Link to this topicShould exceeding a datasheet limit automatically disqualify a part when the circuit still meets its requirements? How should measured parametric drift enter circuit analysis, including temperature, aging, startup, and operating modes? The way failure is defined may carry conservatism of its own.
Link to this topicWhat quantity and kind of test data an acceptance decision requires, and whether the requirements we write are measuring the risk we care about.
A susceptibility screen, an upper bound on event rate, a fitted cross-section curve, and qualification of a mitigation are different decisions with different evidence needs. How many LET or energy points, how much fluence, and how many observed events does each require? Stopping rules and the interpretation of zero observed events belong in this discussion, as does the published work on bounding single event latchup rates from limited data and historical behavior.
Priority discussion: Sufficient SEE evidence and LET requirements
Link to this topicMore fluence on one device and more independently sampled devices answer different questions. When can historical data or process knowledge reduce sampling? When do outliers, mixed die revisions, or poorly controlled commercial populations prevent that reduction? Sample size deserves its own discussion alongside the number of irradiation points.
Link to this topicWhat does passing a single LET requirement establish about mission risk? Ion range, energy, LET at the sensitive region, angular response, and the limitations of cosine scaling for nonplanar structures all affect the answer. When does a threshold requirement provide a useful screen, and when is rate analysis needed instead?
Priority discussion: Sufficient SEE evidence and LET requirements
Link to this topicSEL occurrence, immediate destruction, successful recovery, and subsequent reliability are separate questions. What evidence establishes that detection and power interruption protect the device across all relevant power paths? Repeated events and latent damage must be part of the assessment.
Priority discussion: SEL mitigation
Link to this topicFor single event burnout, single event gate rupture, and related failure modes, when can a measured operating envelope or a bounded failure probability support acceptance? How do voltage, temperature, particle conditions, and device variability affect that conclusion? This discussion is kept separate from recoverable SEL.
Link to this topicWhen evidence taken at the system or board level can relax, replace, or stand in for component-level requirements.
When can error correction, scrubbing, watchdogs, redundancy, filtering, or power cycling relax component requirements? What demonstrates coverage of the relevant failure modes? What does fault injection establish, what still requires irradiation, and how does recovery time affect availability and mission performance?
Link to this topicWhich acceptance decisions can a representative board test support? Component exposure, diagnostic coverage, operating modes, accumulated TID and displacement damage, and the ability to identify the source of a failure all limit what a board test can establish. Where does testing the application reduce uncertainty enough to avoid separate characterization?
Link to this topicWhen existing evidence applies to a new design, and whether component-level acceptance survives aggregation across a bill of materials, a mission duration, or a constellation.
What makes vendor data, another project's test, or flight heritage applicable to a new design? Die identity, fabrication process, package, operating conditions, and mission environment each matter. What changes should trigger retesting? How much confidence can flight success provide when anomalies and exposure are incompletely recorded?
Link to this topicWhen is NIEL scaling sufficient, and when does the device response require more representative testing? When should prior TID, displacement damage, or aging enter SEE qualification? Conversely, when does requiring combined-effects testing add cost without changing the acceptance decision?
Link to this topicHow do small per-device probabilities accumulate across a large bill of materials, a long mission, or a constellation? Shared power and recovery resources, storm-time event rates, and common dependencies in redundant systems all affect the aggregate. When does replacement capability justify relaxation, and when does a shared vulnerability undermine that argument?
Link to this topicThe list above is deliberately incomplete. If an acceptance decision you face is not represented, or if you think one of these questions is asked the wrong way, submit it. Proposals received by 15 January 2027 will be considered for the program.