Real projects must be studied through observable cases and transparent reconstruction.
In this domain, controlled experiments are structurally unavailable in most practical settings. SELRA therefore treats evidence architecture as part of the research program from the start.
- 01
Observational constraint
Industrial projects are observed through real records, implementation histories, and operational traces rather than clean experimental controls.
- 02
Observable cases
Cases can reveal selection and realization mechanisms when boundaries, assumptions, and missing evidence are stated carefully.
- 03
Counterfactual reconstruction
Reconstruction should make alternatives, baselines, completion assumptions, and uncertainty visible rather than hidden in a single result.
- 04
Sensitivity and robustness
Sensitivity analysis helps show which claims survive reasonable changes in assumptions and which remain provisional.
- 05
Evidence-status discipline
Records should distinguish observed facts, reconstructed quantities, assumptions, disputed evidence, and unavailable evidence.
- 06
Confidentiality boundary
Confidential or identifiable project evidence should not be sent without a separate arrangement and appropriate consent.