KML Center Consortium: Problem Summary Report

Document Sample
KML Center Consortium: Problem Summary Report Powered By Docstoc
					                                                                                 KML-PSR-2-Feb10
                 Knowledge Management Laboratory (KML)
                 723 Seibert Rd, Suite 3 Phone: (618) 744–9848
                 Scott AFB, IL 62225     E-mail: kml@cadrc.calpoly.edu




               KML Center Consortium: Problem Summary Report
      Research Request #: 2
          Research Topic: Data Quality
            Requested by: USTRANSCOM-J6
         Submission Date: 9 February, 2010

 1. Description and Research Questions
Lack of data quality continues to be a perennial problem for USTRANSCOM in respect to the
following six desirable data characteristics: accuracy; completeness; consistency; timeliness;
uniqueness; and, validity. Identified problems include stovepipe systems, redundant data storage,
missing and incorrect data values, non-compliance with standard data models, and non-
standardized business processes. USTRANSCOM-J6 is requesting technical advice from the
KML Center Consortium on the following research issues:
 1.1 Identification of the root causes of the data quality problems, investigation of the
     solution issues involved, and recommendation of strategies for addressing these
     problems.
 1.2 Exploration of existing and emerging data cleansing and data mapping
     methodologies, including demonstration of selected capabilities applied to
     representative data in use-cases relevant to USTRANSCOM.
 1.3 Recommendation of steps that can be taken by USTRANSCOM immediately to
     mitigate at least some of the data quality problems.


 2. Typical Use-Case
USTRANSCOM-J6 has provided two use-cases. Use-case (1) deals with surface vessel payload
departures and arrivals (Attachment A). Use-case (2) deals with the special data classifications
that apply to the shipment of hazardous material (Attachment B).


 3. Current Strategies and Procedures
Processes are currently modeled and executed through webMethods and DataFlux is typically
used to profile large quantities of historical data.

 4. Data Requirements
Both the analysis of the underlying data quality problems and the demonstration of selected
capabilities will require access to representative data.



                                                        1
                                                                                       KML-PSR-2-Feb10
                    Knowledge Management Laboratory (KML)
                    723 Seibert Rd, Suite 3 Phone: (618) 744–9848
                    Scott AFB, IL 62225     E-mail: kml@cadrc.calpoly.edu




    5. Operational Environment
The data quality problem appears to pervade virtually all operational processes within the
Distribution Process Owner’s (DPO) operational environment. In particular it impacts the ability
of planners and executors “… to make real-time decisions, prioritize work, collaborate, and
identify potential problem sets before they evolve”1.


    6. Interfacing Requirements
This research project does not require any electronic interfaces to USTRANSCOM’s operational
systems. While an interface to the EIL may be desirable it is not an essential requirement for this
research project.


    7. Objectives and Expected Outcome
The overall objective of this research project is to provide technical guidance to
USTRANSCOM-J6 in its efforts to mitigate data quality problems in the short term and
implement strategies for largely eliminating data quality problems in the long term. The outcome
of the research is expected to include:
    7.1 Identification and exploration of the highest priority data quality problems to become
        a specific focus of the research project.
    7.2 Exploration of process management and software-based methodologies that can be
        applied to mitigate the selected highest priority data quality problems.
    7.3 Demonstration of existing software-based methodologies applied to use-cases of the
        selected highest priority data quality problems.
    7.4 Recommendations pertaining to short term and long term measures that may be
        implemented by USTRANSCOM to initially mitigate and eventually relegate data
        quality problems to minor significance as an operational obstacle.


    8. Resources Required
Resources required for this research project include representative data and subject matter
experts from industry, academia and the Government, as follows:
    8.1 Computing and communication facilities of the KML Center and individual
        participating members of the KML Center Consortium.
    8.2 Technical experts on data cleansing and data mapping methodologies drawn from the
        members of the KML Center Consortium.

1
    Quoted from USTRANSCOM’s submission to the KML Center Consortium (J6-AD Proposal: Business Process
    Modeling and data Quality).


                                                           2
                                                                                         KML-PSR-2-Feb10
                   Knowledge Management Laboratory (KML)
                   723 Seibert Rd, Suite 3 Phone: (618) 744–9848
                   Scott AFB, IL 62225     E-mail: kml@cadrc.calpoly.edu



 8.3 Subject Matter Experts (SMEs) on USTRANSCOM’s data quality problems and the
     business processes that underlie the use-cases that will be modeled in detail for the
     experimental analysis, and demonstration of available data cleansing and mapping
     methodologies, provided by USTRANSCOM.
 8.4 Representative data for demonstrations provided by USTRANSCOM.


 9. Project Technical Committee
(To be decided.)




      Attachments:     Attachment A (Use-case (1) – Surface Vessels Payload Departures and
                       Arrivals).
                       Attachment B (Use-case (2) – Hazardous Material Shipments).




                                                          3

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:11/4/2012
language:Unknown
pages:3