ANALYTICAL DATA SERVICES - Notice of Intent to Sole Source - PG Public Services LLC

The Department of Homeland Security/Federal Emergency Management Agency/U.S. Fire Administration/National Fire Data Center (DHS/FEMA/USFA/NFDC) intends to negotiate and award a contract on a sole source basis (IAW FAR 6.302) with PG Public Services, LLC (PGPS) located at 2200 North George Mason Drive #766, Arlington, VA 22207. More than 40 years ago, the 93rd Congress enacted the Federal Fire Prevention and Control Act of 1974 (Public Law 93-498). This legislation directed the establishment of the agency known as the U.S. Fire Administration (USFA). Among the law's directives was that the "Administrator shall operate, directly or through contracts or grants, an integrated, comprehensive National Fire Data Center (NFDC) for the selection, analysis, publication and dissemination of information related to the prevention, occurrence, control, and results of fires of all types." In the legislation, the following functions of the NFDC were delineated: (1) Provide an accurate nationwide analysis of the fire problem; (2) Identify major problem areas; (3) Assist in setting priorities; (4) Determine possible solutions to problems; and (5) Monitor the progress of programs to reduce fire losses. To accomplish these functions, the NFDC is directed to gather, analyze, and disseminate information on: (1) The frequency, causes, spread and extinguishment of fires; (2) The number of injuries and deaths resulting from fires as well as their causes and nature; (3) Property losses; (4) Occupational hazards faced by firefighters including the causes of deaths and injuries; (5) All types of firefighter activities, including inspection practices; (6) Technical information related to building construction, fire properties of materials and similar information; (7) Fire prevention and control laws, systems, methods, techniques, and administrative structures used in foreign nations; (8) Causes, behavior and best method of control of other types of fire, including forest fires, brush fires, underground fires, oil blow-out fires and water-borne fires; and (9) Other useful and applicable information. One of the tools the NFDC established to accomplish its charter was the National Fire Incident Reporting System (NFIRS). NFIRS was established to collect both data on the fire incident and data on the technical components of the fire itself in an effort to better understand the nature of the national fire problem. NFIRS is a voluntary system, and it includes only those incidents reported to the system by fire departments that report to NFIRS. Therefore, the NFIRS dataset is not based on a statistically selected sample, nor is it a complete count of incidents or casualties. NFIRS currently receives data from approximately 80 percent of the Nation's fire departments. Participating fire departments report an average of 21,000,000 non-fire incidents and 1,000,000 fire incidents each year. In addition, other organizations, both public and private, collect supplementary fire-related data, conduct independent fire surveys that are separately analyzed, and provide yet another useful analysis of the Nation's fire problem. Over the years, NFIRS data files have been provided to the public in multiple formats: dBASE (.dbf), Comma Separated Values (.csv) or delimited text (.txt) files. The current file format is delimited text (.txt) and the files are generally queried using SQL, ACCESS, or statistical software packages such as SAS, SPSS or R. The NFIRS is a relational database that requires analysts to link various data modules together based on unique incident id keys in order to analyze the data. For each NFIRS data module, the qualitative data are collected at a very detailed level based on coded values. Because NFIRS consists of a very intricate structure of detailed codes and the fact that it is not based on a statistically selected sample, NFIRS data analysis is very complex and requires extensive knowledge of the system and its data. NFIRS has unique analytic issues that apply only to this system. For these reasons, it takes several years for analysts to learn the complexities of the system. NFIRS data can present analytic challenges for many fire topics studied. In addition to the complexity of the system, NFIRS has numerous data quality issues that also present additional analytic challenges. NFIRS participation is completely voluntary and data entry is performed by fire departments and personnel that may have limited NFIRS training and system knowledge. Due to this, there are many missing or conflicting values in the data which require the use of sophisticated analytic techniques when analyzing the data. Part of the data analysis includes devising methods to deal with the data quality problems as they arise such as missing and unknown data and invalid values. With the goal of creating a single, consistent, authoritative source of national summary fire data, the USFA/NFDC determined a need to develop a new, transparent, fire incident estimation methodology using public data sources, along with NFIRS data. As a result, Contract No. GS10F0195X, Task Order No. HSFE20-12-F-0240 and Order No. HSFE20-17-P-0268 were awarded and focused on the task of developing a new fire incident estimation methodology. The data analyses performed as part of these contracts required: (1) analysis of large, relational databases with millions of data records, (2) analysis of databases with a coded variable structure similar to NFIRS (i.e., using qualitative data analysis methods), (3) analysis of data that are not based on a statistically selected sample, (4) analysis of data with various data quality issues common to standard databases used in all national fire data analyses, (5) develop new statistical and analytic methodologies for analyzing fire data, and (6) develop imputation techniques to handle unknown and missing data. It was anticipated that the new methodology would replace the existing one used by the USFA/NFDC to establish national annual fire incident estimates of fires, fire deaths, fire injuries and dollar loss. The new fire incident estimation methodology will be completed by the end of July, 2019. The new contract/order is a logical follow-on to work performed under Contract No. GS10F0195X, Task Order No. HSFE20-12-F-0240 and Order No. HSFE20-17-P-0068. The work performed under these past contracts/orders focused on the analyses of fire related incidents. The intent of the upcoming new contract/order is to focus on developing a new non-fire incident estimation methodology with the goal of establishing a new and previously unpublished annual U.S. national estimate of non-fire incidents to which fire departments respond. Non-fire incidents include EMS, EMS/rescue, haz mat, false alarms, service calls, etc. The contractor must be able to meet the core responsibilities for this requirement as follows: 1) analyze large, relational databases with tens of millions of data records, 2) analyze databases with a coded variable structure similar to NFIRS (i.e., using qualitative data analysis methods), 3) analyze data that are not based on a statistically selected sample, 4) analyze data with various data quality issues common to standard databases used in all national fire and non-fire data analyses, 5) develop a national non-fire estimation methodology using the national fire estimation methodology developed in the previous contracts/orders as a basis. 6) develop new statistical and analytic methodologies for analyzing non-fire data, and 7) develop imputation techniques to handle unknown and missing non-fire data. Additionally, the contractor: 1) personnel must have a minimum of 5 years' experience working with NFIRS data and demonstrate that they have an in-depth knowledge of NFIRS and a deep understanding of how the data can and cannot be used. 2) must have a fully developed analytical database consisting of 10 years of National Fire Incident Reporting System (NFIRS) data of valid NFIRS incidents. 3) must have undertaken an exhaustive literature search in peer-reviewed journals, PhD theses, and published reports to identify factors which may influence outcome of fire or non-fire incidents. 4) must have undertaken an exhaustive data sources investigation to identify alternate sources of data for casualties from incidents responded to by fire departments. 5) must demonstrate that it understands and can replicate the methodology used by USFA to extract data from the National Center for Health Statistics (NCHS) Mortality data set. 6) must have working data correction models to account for fire departments' incomplete fire or non-fire incident submission. These models must be immediately available to run on prior year NFIRS data and on the current NFIRS data when that data is available. 7) must have the capability to identify unique NFIRS (fire or non-fire) incidents for analyses and estimates, including mutual aid incidents reported by an aid-giving department for which the aid-receiving department did not participate in NFIRS that year. This capability must be immediately available to be applied on prior year NFIRS data and on the current NFIRS data when that data is available. 8) must have the capability to account for likely numbers of fire or non-fire incidents in geographic areas where no incident data was submitted to NFIRS. This capability must be immediately available to be applied on prior year NFIRS data and on the current NFIRS data when that data is available. 9) must be able to account for known fire departments who failed to submit incident data in a given year. This capability must be immediately ready to be applied on prior year NFIRS data and on the current NFIRS data when that data is available. 10) must have a peer-reviewed paper, suitable for publication that outlines a detailed method for estimating annual fire incidents that may be used as the basis for a method to estimate annual non-fire incidents. 11) must have working models to estimate the numbers of casualties from annual fire incidents that may be used as the basis for estimating the numbers of casualties from annual non-fire incidents. These models must be immediately available to run on prior year NFIRS data and on the current NFIRS data when that data is available. 12) must have working models that identify the key data elements, ancillary data, and their inter-relationships to produce preliminary estimates of the direct dollar loss associated with an incident where appropriate. These models must be immediately available to run on prior year NFIRS data and on the current NFIRS data when that data is available. Based on the above information, the only company known which can provide technically acceptable services for the National Fire Data Center is PGPS. PGPS has an in-depth knowledge of the NFIRS system and its data. They have a thorough understanding of how the data can and cannot be applied. They have the ability to: perform data analyses tasks such as analyzing large, relational databases with millions of data records; analyze databases with a coded variable structure similar to NFIRS (i.e., using qualitative data analysis methods); analyze data that are not based on a statistically selected sample; analyze data with various data quality issues common to standard databases used in all national fire data analyses; and most importantly, they have an intimate understanding of the standard national estimates methodologies currently in use by USFA and have been key in developing new national estimates methodologies. PGPS has demonstrated extensive experience working with NFIRS data; performing data analyses; producing analytic reports and integrated survey sample results that add to the body of knowledge of the fire situation in the U.S. PGPS has a history of performing comprehensive statistical research and detailed fire data analyses of NFIRS data in conjunction with various other data sources, including but not limited to U.S. Census Bureau data, National Fire Protection Association data, National Center for Health Statistics data, and other data sources. FEMA has determined that PGPS is uniquely qualified to perform this work, possessing the expertise and qualifications necessary to complete the required tasks. PGPS performed and completed work under Contract No. GS10F0195X, Task Order No. HSFE20-12-F-0240 and Order No. HSFE20-17-P-0268. Based on their innovative proposed approach and proprietary analytical techniques and methods used, the Government has determined PGPS is the only candidate capable of developing the methodology. This notice is NOT a request for proposal (RFP) for competitive proposals and no solicitation exists for this requirement. A determination by the Government not to compete based on responses to this notice is solely within the discretion of the Government. Information received will be considered solely for the purpose of determining whether to conduct a competitive procurement. FEMA will NOT be responsible for any costs incurred by interested parties in responding to this notice of intent. Only written responses will be considered. All interested parties shall submit clear and convincing documentation demonstrating their capabilities to satisfy the requirements listed above. Submit documentation (using PDF, MS Word, Excel attachments) to Cindy Adams, Contract Specialist at Cindy.Adams@fema.dhs.gov no later than 2:00 P.M. EST on March 13, 2019. Fax inquiries will not be accepted. The capability documentation should include general information and technical data describing in detail the prospective contractor's capabilities and experience that will demonstrate the ability to perform the same services as described in this notice. All capability statements received by the closing of this notification of this synopsis will be considered by the Government. Primary Point of Contact: Cindy Adams Contracting Office Address: DHS/FEMA/NETC Building D 16825 South Seton Avenue Emmitsburg, MD 21727 Place of Performance: Emmitsburg, MD 21727 Classification Code: B506 NAICS Code: 541611