This is a Sources Sought Synopsis. THERE IS NO SOLICITATION AT THIS TIME. The U.S. Nuclear Regulatory Commission’s (NRC) is conducting a market survey and analysis to determine the range of available contractors that exists and to assess their capabilities to assist the NRC in performing the scope of work described below. This request for capability information does not constitute a request for proposal. The NRC does not intend to award a contract on the basis of this request or to otherwise pay for the information solicited.

Jeffrey R. Mitchell, Contracting Officer, at Responses may also be mailed to:
Phone: (301) 492-3639
Fax: (301) 492-3437

U.S. Nuclear Regulatory Commission
Division of Contracts
Attn: Jeffrey R. Mitchell; M/S: CSB-C6D20
11555 Rockville Pike
Rockville, MD 20852

The applicable North American Industry Classification System (NAICS) code assigned to this procurement is 541519. The anticipated period of performance will be 24 months.

The objective of this scoping study is to better assess the feasibility and potential value of advanced knowledge engineering tools, i.e., Content Analytics (principally existing software programs, but also including broad technologies) in supporting risk-informed decision making at the NRC. The study will assess the risk information needs of prototypical agency users of that information, characterize current methods employed by users to meet these needs, identify areas for potential improvement, identify and characterize knowledge engineering tools that may be able to address these areas, perform demonstration analyses using selected knowledge engineering tools, evaluate the results and lessons from the demonstration analyses, and provide an overall assessment of potentially fruitful areas for further development that would result in improved tools for NRC staff use.

This scoping study will explore the use of advanced knowledge engineering (KE) tools and techniques (e.g., natural language questions and answers, text mining, formal modeling) in encoding and applying probabilistic risk assessment (PRA) expert knowledge in PRA document review and PRA system model review. This scoping study will determine if these KE techniques are sufficiently mature to provide NRC staff with (1) ready access to a much larger information base than the PRA and hard-linked documents (i.e., the information base would likely include documents not identified by the PRA document authors or directly related to the PRA field but relevant to the technical issues being reviewed), and (2) flexible, expert-informed tools to query this information base. This project is needed to address the agency’s ever-increasing use of risk information, coupled with the agency’s ongoing and projected loss of risk experts. This project could enhance the efficiency of NRC review of PRA applications. It is anticipated that this technology would assist the staff in connecting diverse sources of information (e.g. PRA documentation, licensing basis information, operating experience, licensee submittals, inspection reports, generic communications) in an integrated manner. This is particularly useful for PRA applications when subtle dependencies and interactions can drive risk results. Moreover, the basic underlying technology could be useful in non-PRA applications that require intelligent mining of large amounts of information.

Content analytics is the act of applying expert intelligence and specialized analytics practices to digital content. Both the private and public sectors have begun to use content analytics software to provide visibility into the amount of content that is being created, the nature of that content and how it is used.

An organization produces two types of content: structured and unstructured. Structured content typically resides in a database. Unstructured content can be found throughout the organization. It can be text-based, as in the case of emails, office documents and Web documents — or non-text-based, such as voice, images or video. Content analytics software uses natural language queries, trends analysis, contextual discovery and predictive analytics to identify patterns and trends across an organization’s unstructured content. Content Analytics (Text Analytics (*1*) and Text Mining (*2*) ) refers to the text analytics process plus the ability to visually identify and explore trends, patterns, and statistically relevant features found in various types of content spread across various content sources. For the NRC, unstructured content includes, but is not limited to, inspection reports, safety evaluation reports, license amendment requests, Commission papers, safety analysis reports, scientific and engineering literature (e.g., journal articles and conference proceedings), and technical reports (e.g., NUREG reports). Structured content includes information contained in operating experience databases (e.g., licensee event reports, accident sequence precursor database). It is highly desirable that any content analytics software used for NRC applications would be capable of utilizing the NRC’s Agency Documentation And Management System (ADAMS) as well various technical databases.
(*1*)Text Analytics describes a set of linguistic, statistical, and machine-learning techniques that allow text to be analyzed and key information extracted for business integration.
(*2*)Text (Content) Mining: extract meaningful data from unstructured text.

The goal of this scoping study is to use content analytics in an effort to gain new insights for improved PRA modeling and improved support of risk-informed decision-making. Potential applications might include:
– Monitoring changes in plant risk profiles,
• Identifying common or generic issues across a range of nuclear plant types,
• Identifying performance trends,
• Identifying and analyzing common PRA modeling elements across a range of nuclear plants,
• Screening, categorizing, and analyzing operating experience data.

Types of advanced knowledge engineering tools and techniques that NRC may acquire from a contractor in the future may include one or more of the following Content Analytic tools or techniques described below.
1. Content Analytics: A range of search and reporting technologies which can provide similar levels of business intelligence and strategic value across unstructured data to that conventionally associated with structured data reporting.
2. Content assessment: Trawling of stored documents and content to measure relevancy, currency, or frequency of access as an indication of the need to keep, or more particularly, migrate content to another system.
3. Content de-duplication: Exact or near-exact match of content stored within the same or different systems, albeit with different metadata as to who stored it and when. Scoring system allows automatic deletion of duplicates, saving space and reducing potential errors.
4. Digital Asset Management (DAM): Content management systems particularly geared up for rich media files such as images and sound which are characterized by large file sizes, proxy representations (low resolution thumbnails or clips) and complex coders, decoders or format transformers.
5. Faceted search tools: Ability to sub-divide within search results by a standard set of metadata tags, e.g., as used by shopping websites to subset a product search by manufacturer, size, color, price range, etc.
6. Federated search: Ability to interrogate more than one repository or index from a single search screen. May include linking internal ECM systems with subscription access to government databases or those of professional bodies.
7. Image and sound tagging: Pattern recognition search to match and apply additional metadata, e.g., faces, trees, voices, birdsong, to rich media files. May also be part of digital forensics.
8. Rich media: Generally used term for non- text formats such as photo-images, graphics, video, sound and animation. They cannot be searched by their textual content so must be tagged and/or represented by thumbnails or audio samples.
9. Sentiment analysis: Analysis of words used in comment or feedback to indicate satisfaction or dissatisfaction with products or services, aggregated to an overall score of satisfaction. Often extended to overall brand response, and monitored for trends or incidents.
10. Text analytics: Lexical analysis to study word frequency distributions, pattern recognition, tagging/annotation, information extraction, data mining, etc., frequently as part of other processes described here.
The contractor shall comply with all applicable Federal Information Processing Standards (FIPS).
All commercial software provided for the purposes of Content Analytics shall comply with the Unstructured Information Management Architecture (UIMA) Version 1.0 OASIS Standard, 2 March 2009 or other applicable standard(s).

The contractor may be required to be certified under Federal Risk and Authorization Management Program (FedRAMP) (see and may be required to comply with the security directives, procedures and controls as stipulated in FISMA, the National Institute of Standards and Technology (NIST) 800.53 publications, and NRC IT security policies and controls. Please describe where you are in the FedRAMP process.

Capability Sought
The NRC is seeking to identify commercial organizations capable of providing technical assistance in the use of Content Analytics tools and techniques (e.g., natural language questions and answers, text mining) in encoding and applying probabilistic risk assessment (PRA) expert knowledge in PRA document review and PRA system model review.

Mandatory Qualifying Criteria
The contractors should be technical experts in their field, with extensive experience in applying expert intelligence and specialized analytics practices to digital content using Content Analytics. Membership in professional associations and active participation at relevant professional meetings such as the Organization for the Advancement of Structured Information Standards (OASIS) conferences is desirable. The contractors should have in-depth knowledge and corporate and individual experience and expertise in their technical area(s) in order to identify and resolve issues. The primary Content Analytics software proposed must comply with the latest version of the OASIS Standard for Unstructured Information Management Architecture (UIMA). Corporate experience applying Content Analytics with US Government agencies is desirable.
Commercial organizations that are interested in supporting our technical assistance requirements please address the market research questions below and provide capability information on your staff’s qualifications and your firm’s corporate experiences and qualifications on similar contracts or efforts.

Market Research Questions
If your organization has the potential capacity to perform these contract services, please provide the following information:
1) Organization name, address, email address, website address and telephone number?
2) How long has your company been in operations?
3) What size is your organization with respect to NAICS code 541519 (i.e. "small" or "other than small")? If your organization is a small business under the aforementioned NAICS code, what type of small business (i.e. small disadvantaged business, woman-owned small business, economically disadvantaged woman-owned small business, veteran-owned small business, service-disabled veteran-owned small business, 8(a), or HUBZone small business? Specify all that apply.
4) How many people does your company employ, including consultants? Please break down the mix between the two categories.
5) Describe teaming arrangements your company has formed to perform scopes of work outside your core competencies. Describe both the type of work and how you managed it.
6) Which GSA Multiple Award Schedules or Governmentwide Acquisition Contracts, if any, does your organization possess? When do they expire?
7) Does your organization have a Defense Contract Audit Agency (or other cognizant audit authority) approved accounting system?
8) Does your organization have agreed upon indirect rate agreement with any government agency?
9) Has your company previously faced organizational conflict of interest issues with NRC? How were they resolved?
10) List any companies or government agencies your company either plans to or has performed work for related to civilian nuclear reactors. Describe the work performed.
11) Although no geographic restriction is anticipated, if responding organizations are located outside the Washington Metropolitan area, indicate how the organization would coordinate with the NRC program office located in Rockville, MD.
12) Describe the experience your organization has with respect to the objective above. Include which of the ten (10) types of tools or techniques listed above that your organization has experience providing as a prime contractor.
13) For any federal customers that your agency has for items 1) through 10) above, indicate how your company complies with applicable Section 508 standards (see
14) Indicate whether your company is certified under FedRAMP and, if not, indicate whether your company has applied for FedRAMP certification, what stage in the FedRAMP certification process the company is in and whether the company can project an anticipated date for becoming certified. If the company has not applied for FedRAMP certification, indicate whether the company has a plan to apply for FedRAMP certification and, if so, when the company plans to apply for it.
15) Provide a tailored capability statement addressing the particulars of this effort, with appropriate documentation supporting claims of organizational and staff capability. Organizations responding to this market survey should keep in mind that only focused and pertinent information is requested. If significant subcontracting or teaming is anticipated in order to deliver technical capability, organizations should address the administrative and management structure of such arrangements. Taking into account the magnitude of the scope of this effort, organizations also should address the capacity of their financial infrastructure to coordinate and deliver contract performance.

Government Evaluation
The Government will evaluate market information to ascertain market capacity to:
1) Potentially provide the services consistent in scope and scale with those described in this notice and otherwise anticipated;
2) Potential capacity to secure and apply the full range of corporate financial, human capital, and technical resources required to successfully perform similar requirements; and
3) Potential capability to implement a successful project management plan that includes: compliance with tight program schedules; cost containment; meeting and tracking performance; hiring and retention of key personnel; and risk mitigation;
The purpose of this announcement is to provide potential sources the opportunity to submit information regarding their capabilities to perform work for the NRC free of Organizational Conflict of Interest (OCOI). For information on NRC OCOI regulations, visit NRC Acquisition Regulation (NRCAR) Subpart 2009.5, entitled "Organizational Conflicts of Interest" (

All interested parties, including all categories of small businesses (small businesses, small disadvantaged businesses, 8(a) firms, women-owned small businesses, service-disabled veteran-owned small businesses, and HUBZone small businesses) are invited to submit a response to the market research questions below and submit the capability information as described below. The capabilities package submitted by a vendor should demonstrate the firm’s ability, capability, and responsibility to perform the principal components of work listed below. Submission of additional materials such as glossy brochures or videos is discouraged. Responses are due no later than August 16, 2013. The Government will not reimburse respondents for any costs associated with submission of the requested information. Telephone inquiries or responses are not acceptable. Responses should be no longer than 25 pages in length. Text in the main body of the response must be a font size of 10 or larger. In addition to the 25-page response, applicable pre-printed marketing material may also be included with the response.