AI tapped to streamline past performance evaluations
- By Anoushka Deshmukh
- Aug 14, 2019
It's not news that the acquisition community wants to improve the Contractor Performance Assessment Reporting System. For vendors, these evaluations are often the thin line between acquiring or losing a contract. For agencies, CPARS contains so much data that it's difficult for contracting officers to quickly pull relevant information.
As the official source for past performance data, CPARS includes information regarding a contractor’s actions under previously awarded contracts as well as its conformance to requirements and standards of good workmanship. It also contains data on the vendor's forecasting and controlling costs, adherence to schedules, commitment to customer satisfaction and business ethics -- along with other information.
In an Aug. 13 solicitation, the Department of Homeland Security called for an artificial intelligence-based tool that can help contracting officials rapidly pull relevant information on previous evaluations for a given contractor or vendor. This tool would help streamline the federal contracting process and alleviate large administrative workloads.
DHS seeks a proof of concept/prototype of the tool that uses advanced automation, continual learning, prediction and decision aids to solve complex, dynamic problems. It also requires the AI be "human readable," meaning humans can understand how the AI delivers results, as well as verifiable, meaning it produces objective results "that can be trusted, are free from biases, and that produce intended outcomes in a repeatable and objective manner."
Proposed solutions should not burden contracting officers. Ideally, the solicitation said, the solution would learn about a solicitation through machine reading, rather than having the contracting officer input the data. Even better would be a way for contracting officers to input only a DUNS number to start the proposed solution’s processes.
Besides using AI to streamline CPARS evaluations, DHS also wants data-driven recommendations about improving the quality of past performance data entered into CPARS by contracting officers.
The government will award up to $50,000 to vendors who can provide a prototype that successfully delivers relevant information based on test data. Follow-on contracts may be awarded to prototypes that use AI to "predict whether the Offeror being evaluated will successfully perform the future contract based on its past performance record in CPARS, without further competition."
The ultimate goal of the CPARS AI initiative would be a "self-sustaining multi-contractor commercial marketplace, perhaps similar to the credit report market place," the solicitation said.
Read the full notice here.
Anoushka Deshmukh is an intern with Public Sector 360, writing for GCN, FCW and Defense Systems.