In This Section
4C Partners
NASA Cost Estimating Tool (NASA-CET)
| Property | Description |
| ID | 2 |
| Name | NASA Cost Estimation Tool (NASA-CET) |
| Creator & Funding | Developed for NASA by SGT (Stinger Ghaffarian Technologies, Inc.) |
| Status | The first version of CET (version 1) was published in 2004. The newest version available (version 2.4) is from September 2008. |
| Purpose | Estimating life cycle costs for ground data centres activated to improve budgets for NASA missions. |
| Information assets | Space data - typically multidimensional data sets |
| Activities | Ingest, Data Management, Archival Storage, Access, Administration |
| Resources | Capital cost (system purchase, maintenance, commercial off the shelf (COTS), source lines of code (SLOC), facility, media), labour cost (5 levels). |
| Time | Past, Present, Future - for a 7 to 10 year time scale, reflecting normal data processing time period for missions. |
| Variables | 96 distinct descriptors, such as staff salaries, system purchase cost, COTS software license, archive media, inflation, volume, automation level. |
| Type of tool | Analysis, estimation and review of estimation. Implemented in MS Excel Spreadsheet using Visual Basic (VBA) |
| Availability of tools | The CET tool, user guide, technical description etc is available for download. |
| References | Fontaine, K., Hunolt, G., Booth, A. and Banks, M., Observations on Cost Modelling and Performance Measurement of Long Term Archives, in PV2007 Conference Proceedings, 2007 Hunolt, G., Booth, B. and Banks, M., Cost Estimation Toolkit (CET) Users' Guide, Version 2.4, 2008 (available as part of the CET software package) |
The NASA-CET (CET) is designed to generate life-cycle cost estimates for implementing, operating and maintaining a science data system. It employs the cost estimation by analogy approach, using information about existing data activities as the basis for estimating life cycle-costs for user-defined activities. It is intended for use on science data, but can be applied to other institutions. The CET divides the life-cycle into a set of functional areas that cover different generic areas of costs.
These compose the reference model behind the tool. The functional areas include: ingest, processing, documentation, archive, access and distribution, user support, sustaining engineering, engineering support, technical coordination, implementation and facility /infrastructure. Based on the functional areas users can describe activities and the CET then estimates the costs for the user-described activity and outputs the results in spreadsheet and graphical formats. The effort is expressed in Full Time Equivalents (FTEs), which is equivalent to that of a single person working full time for a year. With this model and its tool a user can enter their own data set for a new similar activity, estimate its cost, and review it.
The CET uses regression to develop the coefficients for a set of seven trial relationships of FTE to workload parameter for each of the selected workload parameters.
|
Linear |
Y = a + b x X |
|
Logarithmic |
Y = a + b x lnX (ln is natural logarithm) |
|
Exponential |
Y = a x e (b+X) (e is the base of the natural logarithm) |
|
Quadratic |
Y = a + b x X + c x X2 |
|
Square Root |
Y = a + b x X + c x √X |
|
Linear-Logarithmic |
Y = a + b x X + ln(X) (ln is natural logarithm) |
|
Linear-Exponential |
Y = a + b x X + c x eX (e is the base of the natural logarithm) |
For the first three relationships, the CET uses single parameter regression of Y’s on X’s (effort on workload), and for the last four, two-parameter multiple regression; for example, for the quadratic case the two parameters are X and X2.
The estimate is by analogy, in other words the data sets come from similar NASA missions with similar activities. A tool has been made to support the import of the data sets to the comparable databases.
The model has a data activity reference model. OAIS is used as a reference for the model (a mapping has been made). Note that OAIS preservation planning and migration activities are not explicitly included in the model and its tool. CET does not directly address long-term archiving; it would need extensions to be able to do that.
The NASA-CET model covers a lot of detailed information for in-project costs so could be very useful for Principal Investigators (PIs) wanting to assess Research Data Management (RDM) and sharing costs for new research projects. However, the learning curve is probably too steep and the amount of time needed to capture and model the activities to enter into the tool may require too much for most researchers to want to use it. The level of complexity in the tool may be usual for the space sciences but for most other disciplines it is far too complicated. If it were simplified greatly though, it could be very useful as a project costing tool.
