The RFP Database
New business relationships start here

SOURCE CODING FOR DATAP


California, United States
Government : Federal
RFP
Go to the link
This document has expired, therefore the above link may no longer work.

NAICS: 541511 Custom Computer Programming Services / Size Standard $27.50
This is a Request For Information B? This synopsis is for information and planning purposes only and is not to be construed as a commitment by the Government. No contract will be awarded as a result of this Sources Sought announcement. The United States Geological Survey (USGS), Office of Acquisition and Grants (OAG) is soliciting information from qualified sources; the results of this announcement will be evaluated to determine if there are businesses capable of performing the proposed work.
The USGS, OAG seeks responses from small business vendors that possess the capability to conduct technical, professional, statistical, and programming services for the USGS, Development of Assessment Techniques and Analysis Project (DATAP) funded by the USGS Mineral Resources Program.
Custom software to perform the simulation was developed by David Root in the 1980s to run on a Data General computer (Root and others, 1991). An Apple Macintosh version was released in 1998 (Root and others, 1998) and PC versions in 2000, 2004, and 2012 (Duval, 2000, 2004, and 2012). New and future releases of the Windows and Apple operating systems will ultimately make it impossible to run the existing code. The USGS scientists who wrote the code and ported the software to different computer platforms have retired from the USGS. Because of poor documentation, no one fully understands how the source code works or knows how to make modifications.
We need to reproduce the basic functionality of the simulation software programmed by Root using a coding platform that is well-documented, modular, and open-source B? free to distribute and not tied to features in operating systems that will require constant updating. We also need to have access to the original computational approach so we can compare new results with those done in the past.
The contractor shall work with the project chiefs and project members by providing scientific, technical, and statistical assistance in evaluating our needs for undiscovered resource estimation, designing a computational approach that will allow us to continue using expert estimation of undiscovered deposits as an assessment technique, and creating and documenting code using the R statistical computing package to do Monte Carlo simulation in support of our assessments of undiscovered mineral resources. .
Technical Requirements of Contractor
The job requires (1) a thorough understanding of energy and mineral resource assessment methodologies of various types as used by the USGS, (2) an advanced knowledge (Ph.D. level or higher) of probability and statistics, as applied to geological resource issues, (3) a broad background in basic, applied, and pragmatic statistical approaches and methods, and (4) expertise in R programming.
Criteria for acceptance include a demonstrated knowledge as indicated by (1) a publication record indicative of both superior mastery of the concepts of probability and statistics as applied to earth science problems and (2) specific knowledge of and familiarity with USGS assessment methods and practices.




Work Requirements
Reproduce the simulation approach published by Root and others (1991) using the open-source R statistical computing package.
As originally described by Root and others (1991), the proposed code should:
B?    Use the algorithm developed by Root to convert expert estimates of undiscovered deposits (subjectively estimated at various quantiles) into a distribution that can be used in the simulation.
B?    Allow for the user to specify the probability of zero deposits.
B?    Address the dependencies between ore tonnages and grades of deposits AND between grades of different metals in the same deposit type.
B?    Have both a lognormal and empirical option for modeling deposit ore tonnage and grade distributions.
B?    Use the same approach by Root to deal with missing grade data. The missing grades are assumed to be zero and the input dataset is divided into B?metal suitesB?, representing each unique combination of metal associations, in order to calculate covariance in (see Root and others, p. 131 and 132).
B?    The output should retain B?metadataB? that summarizes the information used for the simulation (numbers of undiscovered deposits, the ore tonnage and grade models used, mode of operation, and so on).

For further information, interested parties may contact the Contracting Officer, Charlan Cabalsi at (916)278-9329 or ccabalsi@usgs.gov by the due date, 11/07/2016.

Cabalsi, Charlan

ccabalsi@usgs.gov

    1. Home
    2. Articles
    3. Login or Register

    4. Search

    5. Add/Announce your RFP