Posted by Examination of Functional Correlation And Its Impact On Risk Analysis
Risk Track
Downloadable Files:
RISK-3 Smith Paper Functional Correlation
Abstract:
The term “functional correlation” is often mentioned in papers and presentations, but the definition of the term remains nebulous enough to confound even senior analysts. Yet, functional correlation often appears in even the simplest parametric cost models. Cost estimate methods that share the same uncertain inputs or are simply a factor of another cost method are examples of “functional” correlation. For example, using the same labor rate (with uncertainty assigned) across many different work breakdown structure (WBS) elements causes these WBS elements to be correlated, intended or not. A more profound impact is caused when the common uncertain variable, such as learning slope, has a non- linear impact on the WBS elements that use it. Often, this impact is not recognized or managed properly during uncertainty model development. Further, correlation applied on top of functional correlation can affect not only the variance at the total levels in the model, it impacts the mean. Correlation applied in simulation tools on throughputs (with no functional correlation) can have a significant impact on total model variance, but will have no impact on the mean. Simulation tools are used to replicate published results with respect to correlation on throughputs. From this basis, other simulations investigate the impact of departing from the assumptions upon which the theory is based. This paper describes the difference between functional and applied correlation, identifies how functional correlation is introduced into a model (even if unintended), describes how to measure it and demonstrates, in quantitative terms, the impacts of functional versus applied correlation on the uncertainty analysis.
Author:
Alfred Smith
General Manager for the Software Products/Services Group. Over 20 years experience leading, executing or contributing to life cycle cost model development for a wide variety of systems such as Aircraft Carriers, Submarines, Multi Mission Aircraft, Amphibious Assault Vehicles, International Space Station and Customs Modernization. Over 10 years of experience teaching cost estimating, statistical analysis and risk modeling. Tecolote lead for the development, distribution and support of the ACEIT suite of tools. Education: 1986 University College, London, England, Master of Science with Distinction; 1978 Royal Military College, Kingston, Bachelor of Engineering.