Validation
What the current validation supports — and what it does not yet claim.
Transparency about scope is part of the product. Here is what has been tested, how, and where the claims stop.
01Reference validation
How Qistas has been tested.
- 01
Real public data
Qistas has been evaluated using real public EV operational data.
- 02
Causal pre-decision protocol
The reference implementation only uses evidence available before each decision.
- 03
No future leakage
Future outcome data is excluded from the original assurance decision.
- 04
Deterministic replay
Evaluated decisions are re-run from preserved inputs and must match exactly.
- 05
Duration-aware behavior
Assurance is tested against the duration a commitment requires.
- 06
Evidence validity paths
Stale and infeasible evidence paths are tested, not only the normal case.
02Validation
What the evidence supports — and what it does not yet claim.
What the current validation supports
- Deterministic assurance behavior in the validated reference scope.
- Exact replay of evaluated decisions from preserved inputs.
- Conservative handling of stale and infeasible evidence in tested paths.
- Duration-aware assurance in the reference implementation.
What it does not yet claim
- Enterprise ROI for a specific operator.
- Validation across every DER or asset class.
- Guaranteed delivery of any commitment.
- Regulatory approval or certification.
- Production performance in a customer environment.
03Reference results
Figures, with their scope attached.
Qistas
Fresh unseen validation
Frozen decision rules evaluated on data not used during calibration.
- external publication compression
- 98.7248%
- decision-relevant status transitions captured
- 23 / 23
- ADMISSIBLE → INFEASIBLE transitions captured
- 17 / 17
- downward overstatement events
- 0
- stale publications
- 0
- deterministic replay
- 26 / 26
Results shown are from controlled public-data validation and shadow evaluation. They are not customer-production performance claims.
Qistas
Portfolio validation
Conservative composition of asset-level assurance results into portfolio commitments.
- of internally admissible capacity-time preserved by the conservative assurance layer
- 95.61%
- portfolio overstatement events
- 0
- deterministic portfolio replay in the evaluated dataset
- 100%
Results shown are from controlled public-data validation and shadow evaluation. They are not customer-production performance claims.
Detailed validation reports and methodology are shared during enterprise evaluation, under NDA where required.
04Methodology
How a result earns its place on this page.
- 01
Calibration is not validation
Rules and thresholds are set on calibration data. Results are reported only on data that played no part in setting them.
- 02
Fresh and holdout evaluation
Validation uses unseen periods, assets or cases. Where a holdout exists, it is evaluated once, after rules are frozen.
- 03
Frozen thresholds
Decision rules are versioned and frozen before evaluation. A result is attributed to a specific policy version.
- 04
No outcome-driven rescue
Rules are not adjusted after seeing evaluation outcomes in order to improve them. A changed rule is a new version requiring new evaluation.
- 05
Historical failures preserved
Failed versions and their results are retained. Progress is measured against them, not by replacing them.
- 06
Deterministic replay
Every evaluated decision is re-executed from recorded state and policy. Replay must reproduce the original decision exactly.
- 07
Shadow before control
Live deployments begin in shadow mode. Decisions are computed and compared but not enforced until evidence supports it.
- 08
Evidence lineage and versioned policy
Each decision references the evidence it consumed and the policy version that governed it.
05Deterministic, precisely
Reproducible decisions — not predictions of the future.
With the same evidence, obligations, policy and software version, Qistas produces the same assurance result.
This does not mean that future delivery is guaranteed. It means the assurance decision itself is reproducible and auditable.
- We do not describe controlled benchmark performance as universal proof.
- We do not present shadow economic results as customer savings.
- We do not treat missing evidence as permission.
Enterprise enquiries
Evaluate it on your own evidence.
The most credible validation is the one run on your data. We structure evaluations as shadow studies with pre-agreed metrics.