Description
ALERT System - Request for Information Description: The following Request for Information (RFI) is issued for purposes of conducting market research and allowing commercial IT companies to propose technologies and/or methodologies that could be utilized to identify and monitor potential fraud in the Food Stamp Program. All responses are due not later than 11:00am (local time) Wednesday, October 1, 2003. Comments, suggestions and questions about the proposed new technology may supplement or be incorporated into the United States Department of Agriculture (USDA) Food and Nutrition Service’s (FNS) decision support system called the Anti-fraud Locator using EBT Retailer Transactions (ALERT). FNS is in the process of formulating the acquisition and project strategy for the next base contract and as a result wants to identify commercially available technologies that could be either incorporated into the ALERT system or separately implemented. FNS has obtained the support of Information Systems Support, Inc. (ISS) and MAXIMUS, Inc, as independent technical consultants assisting in the review of the RFI responses. ISS/MAXIMUS is prohibited from submitting a proposal to any procurement request that may result from this RFI. All information submitted in response to this RFI will be used for planning purposes only. FAR Clause 52.215-3 Request for Information or Solicitation for Planning Purposes (OCT 1997) applies. All information will be safeguarded from unauthorized disclosure. Please ensure any sensitive or protected information is clearly marked as such. FNS will consider all information received for future incorporation into ALERT as it evaluates the responses received from the vendor community. This is not a formal solicitation under FAR Part 15 or a FAR Part 13 simplified acquisition, but a request for interested parties to provide information. No solicitation document exists and requests for a solicitation will be unanswered. An overview of data elements and other technical information regarding the current system may be accessed at the USDA website: http://www.fns.usda.gov/. COTS: Vendors with various technology products are encouraged to respond to this request with a high level concept paper that explains the approach and offers new technologies and/or methodologies that would improve the current tools of FNS in its analysis of ALERT data. It is requested that responses include an estimate (Rough Order of Magnitude of the time and costs (hardware, software, and personnel) needed to design and implement. We do not expect full cycle system developments to be offered, but are specifically interested in proven products, technologies and/or tools that enhance or provide additional capabilities to identify and monitor potential fraud using the existing and future data elements available from Food Stamp Program Electronic Benefit Transfer (EBT) transactions. FNS will use the information received from the responses to this RFI to help determine the requirements for the next RFP for ALERT support. FNS anticipates releasing the RFP for ALERT support in the 1st quarter of 2004. Current Architecture: The current ALERT application is a n-tiered server architecture that operates on Proliant 8500 servers using the Windows 2000 operating system and Sybase database system software. The application software is programmed using Delphi. There are four main software components ¨ The ALERT analysis software includes specific scans that target certain suspect patterns of fraudulent activity. ALERT utilizes data from two sources; transaction data that is received from the states operation EBT systems, and retailer data that is extracted from present mainframe systems that house the FNS national retailer database for authorizing and monitoring participating Food Stamp retailers. ¨ A web site provides user information on existing releases, tutorials and a data mart known as AMART. ¨ ESSRA is a component which provides retailer management data that has been extracted weekly from the FNS national retailer tracking database. ESSRA is used in the field and regional offices to manage daily retailer authorization activities and allows users to download data into standard Excel spreadsheets or Access tables for further manipulation ¨ Watchlist is a management reporting tool designed to systemically track identified suspect retailers and the status of activities by FNS staff regarding each identified retailer. FNS’ overall enterprise architecture is transitioning from mainframe to server and web-based technologies. Current software standards require Section 508 compliance for all agency systems. FNS is migrating to SQL Server 2000, which will become the database system for ALERT in the future. Users are provided Windows 2000 and Office Suite products on their workstations or laptops installed with the ALERT application software. Depending on access to the Department Wide Area Network, users connect via the WAN or utilize dial-in capabilities. Future solutions may entail use of Virtual Private Networks (VPN) for secured connections via the Internet. The solution we may deploy would need to be supportable on our present and future system architecture; however, we will not preclude any technological tools should they present an innovative tool that merits consideration in this initiative. Background: ALERT is a decision support management system built around the retailer management and compliance requirements of the Food Stamp Program. The primary purpose is to load and analyze food stamp transactional data, e.g., food stamp purchases in any of 150,000 authorized food stamp retailer locations. Data is submitted to FNS by State EBT processors each month. Only a few States remain to be implemented, the largest being California. At present, there are nearly 70 million transactions processed monthly and a rolling 6 months of data is maintained on-line. When all states have implemented EBT, monthly transaction data is estimated to be about 85 million transactions. Key elements of each electronic transaction are extracted by the EBT processors and provided to the ALERT system. This data includes information such as dollar amount, location, terminal ID, date and time of the transaction, type of entry (card swiped through a reader or manually key –entered), and whether a paper voucher was issued. Business Requirements: FNS is looking for technology and analytic tools that can enhance the current system functionality. The FNS preference is that the system architecture is web and server based. The system must be accessible to a user base of over 600 local and remote users that are accessing data for inquiry, query, and download. Users currently access the system via a wide area network or through dial-up access. Any proposed solution should be able to offer the following: ¨ Ability to identify evolving trends in suspect retailer behavior as they change. ¨ Allow for ad hoc query capabilities ¨ Identify possible relationships that either enhance existing scans and pattern analysis or support the overall retailer management and integrity goals of FNS. ¨ Support set pattern analysis within the FNS systems environment. There is considerable latitude for the range of proposed solutions. These could include COTS solutions, improvement of existing methodologies, or some new technology. The proposed solution is desired to be compatible with existing and planned systems architecture within FNS. However, should a solution not be compatible with said architecture, yet otherwise possess merit and benefit to accomplishing the objectives of FNS, it will be considered. Future Considerations: The existing data analysis being performed within ALERT has undergone limited revision since its inception. The current requirements for data being received from the EBT processors have not been altered since the system was initiated in 1996/1997. There is additional information within the transaction data being collected by the EBT processors that may be of use to FNS, such as telephone nodes or card replacements. Consequently, new data elements may be added in the future to the transaction records being received from the EBT processors. Potential solutions being recommended by the vendor community should take this into consideration. Responses: Vendors interested in responding to this RFI need to submit a capability statement that clearly and convincingly identifies their capabilities. In addition, potential vendors must substantiate their capability with the following supplemental information: methodology, technical data, design, brochures, product names, etc. If these items are available under a GSA schedule, please provide the contract schedule and information. All responses must be received no later than 11:00am (local time) Wednesday, October 1, 2003. Responses must be submitted to the following address: USDA/FNS/ASD/CMB, Tina M Nevitt Contracting Officer, 3101 Park Center Drive Room 228, Alexandria, VA 22302-1500. Your response shall not exceed 15 single sided pages, plus supplemental information. You are required to include an electronic copy in Microsoft Word or a compatible format (Adobe PDF).