New Version

Assembly-line scheduling — QPC on IBM Heron

Three parallel production lines, twelve jobs, one coupled assignment problem (Hernandez et al. case class, arXiv:2605.30252). QPC runs line A / line B / line C as polycontextural blocks in one IBM Heron submission — the recommended customer tryout domain.

Plain English

What this is: Assign each job to exactly one assembly line while balancing processing time, tardiness, and changeover penalties — factory scheduling, not generic QUBO marketing.
What QPC computed: 12-job pilot on ibm_fez · 36 active variables · 156Q Heron layout · 8192 shots · decoder = schedule cost (lower is better).
QPC result: Polycontextural decoder: schedule cost 39.7936 · 0 late jobs · one coupled job · d96daa0tcv6s73djknig.
What that means: Every job lands on a feasible line with no late-job flags under the declared threshold — a concrete operations plan from one quantum run, without classical partition loops between lines.

Lane B · Industry pilot. Headline 0 late jobs uses QPC’s declared problem decoder on quantum samples from one coupled 156-qubit job — an operations KPI, not an ICC architecture witness.

Try your own line weights: optimization showcase · Architecture proofs: JSC.

Customer tryout. Manufacturing buyers map showcase sliders to Line A / B / C priority and coupling between lines — same 8-parameter pattern as this pilot, bounded for evaluation.

IBM Fez — Heron156 (July 2026)

QPC execution	Schedule cost	Late jobs	Qubits	Depth	IBM job ID
QPC polycontextural (one run)	39.7936	0	156	116	`d96daa0tcv6s73djknig`

Three line contextures co-reside on Heron — morphogrammatic padding, transjunction bridges on shared job families, one gate-native Fez submission for multi-line scheduling structure.

QPC circuit layout (one submission)

warm-start assignments → [Line A contexture: local scheduling cost on line-A family jobs] → transjunction (shared sequencing variables) → [Line B contexture: local scheduling cost] → transjunction → [Line C contexture: local scheduling cost] → transjunction closure → one-hot mixer (one line per job) → measure → schedule_cost() + late_jobs()

Assembly-line scheduling — QPC on IBM Heron

IBM Fez — Heron156 (July 2026)

QPC circuit layout (one submission)

Related pilots