Every DLA solicitation.
Priced and reconciled.
The internal console for our DIBBS operation. It pulls every solicitation DLA posts, prices it against real award history, and reconciles what we quoted against what was actually awarded.
Four subsystems, one record.
One resolved view per NSN: approved sources, previously awarded CAGEs, the part-number library out of PUB LOG, and DLA standard pricing. Approved and awarded are kept distinct - a vendor cleared to supply is not the same as one that has won.
Conditional 90/70/50 price bands, quantile-fit and backtested across 33.9k solicitations. Fixed multipliers were tested and rejected - they were wrong in exactly the places that matter.
Every quote is matched back to its award nightly, so win rate and realised value come from the record rather than from memory - and the gap between what we bid and what won is visible per NSN.
RFQs, awards and amendments on a 10-minute poll, plus a day-ahead pull that surfaces tomorrow's solicitations tonight. A nightly parity audit re-asks DIBBS what it has and flags any drift.
What the console does.
Priced from the record, not from memory.
Six upstream sources are reconciled into one view, so a price has an award behind it and a vendor has a history behind it.
