The most useful quantum-computing result is not always a faster answer. Sometimes it is a well-supported decision not to run a quantum workload at all.
That is the central lesson in research developed in response to an E.ON grid-expansion challenge context and documented in the Haivian Research Suite. It is not evidence of an E.ON deployment, access to E.ON operational data or E.ON endorsement. The work began with a practical planning question: where, if anywhere, did a bounded distribution-grid investment problem contain a decision kernel that both mattered to the planning workflow and deserved quantum treatment?
For readers evaluating a future quantum project, the practical takeaway is a five-part admission test: start with a strong classical baseline; treat a valid formulation as one gate rather than a conclusion; count loading, readout and verification costs; keep negative results narrow; and ask whether the target changes a real decision.
The research did not assume that identifying a possible quantum formulation would answer that question. It separated three claims:
- Is the bounded decision target specified well enough to test?
- Is the target difficult for strong classical methods?
- Does a quantum or hybrid route have a credible end-to-end role after loading, decoding, verification and resource costs are counted?
Those distinctions changed the outcome.
No defensible target was admitted
The public quantum-screening view and static evidence summary record the result at the level the public evidence supports: no defensible target was admitted within the tested scope.
The same public summary records zero Qonduit execution calls and no simulator, provider or quantum-hardware run. It also makes no claim of quantum advantage. Those absences are part of the result, not gaps to be filled with a demonstration.
The recorded conclusion was deliberately narrow. It did not say that quantum methods have no possible role in grid planning. It said that the tested scope did not justify admitting one.
What the stop decision protected
The stop protected the research programme from conflating four things that are often blurred together:
- expressing a problem in quantum-compatible notation;
- running a simulator;
- sending a workload for external execution;
- demonstrating useful quantum computation.
None of the screened work earned an execution step. A simulator trace would have shown only that a simulator could execute a constructed object, not that the object deserved quantum resources.
The evidence-led response was to return to problem formulation. Later appraisal asked whether a consequential bottleneck existed inside the broader planning process and whether strong classical methods actually struggled with it. It did not admit a target on the evidence available at the recorded date.
This is a disciplined form of progress: the programme learned which claim it could make, which claim it could not make and what evidence would be needed next.
Why a negative result belongs in a research journal
Quantum research is vulnerable to a simple incentive problem: an execution is visible, while a decision not to run can look like inactivity. But the decision not to execute, disclose data or promote an easy target can be the more technically meaningful result.
The E.ON record turns the admission test into five concrete questions:
- Require a strong classical baseline before a quantum target is admitted.
- Treat a valid formulation as one gate, not a conclusion.
- Count state preparation, readout, reconstruction and verification in the comparison.
- Preserve scoped negative results rather than turning them into universal claims.
- Ask whether the target changes a real decision, not merely whether it fits a circuit.
The result was not “quantum has no role in grid planning.” It was narrower and more useful: this bounded target did not justify one, and the programme should not pretend otherwise.
Evidence boundary
The public evidence was reviewed for this candidate as at 2026-09-28. The quantum-screening view is JavaScript-routed; the static evidence summary is the robust public fallback. Neither is an E.ON operational-data portal, deployment record or endorsement.