Reverse‑logistics operators often must decide inspection depth and recovery routing before an asset’s condition is fully observed, while full inspection consumes scarce labor. Semantic Signal‑Assisted Decision Support converts return notes into a condition factor and a signal‑quality score, guiding inspection depth and recovery allocation under shared labor capacity.
We evaluate the framework in three synthetic benchmark scenarios: IT decommissioning, aircraft maintenance, and consumer‑electronics returns. Each scenario runs 30 paired simulation seeds, comparing a keyword‑based implementation with a noisy full‑inspection structured‑feature baseline. The keyword approach improves net recovery value relative to the baseline and reduces inspection cost across all scenarios. A risk‑blind comparator that skips inspection altogether still records higher value under the benchmark’s purely economic objective.
When inspection cost is matched, score‑guided targeting adds $53.9 k per batch in the aircraft scenario but yields little economic effect in the other two configurations; phrase‑ and large‑language‑model extractors provide additional gains in the aircraft case.
These results demonstrate that narrative evidence can support inspection allocation before recovery decisions are made.
Review