The Daniel Guggenheim School of Aerospace Engineering at Georgia Tech

ROSETTA

Overview

ROSETTApic.png
ROSETTAlogo.png

The Relational Oriented Systems Engineering Technology Tradeoff Analysis (ROSETTA) Initiative is attempting to bridge the gap between qualitative subject matter expert driven techniques and quantitative modeling and simulation techniques. In the same way that the Rosetta stone provided a means to translate between the Greek, Hieroglyphics, and Egyptian demotic languages by having the same text (a decree) repeated in all three languages, the Relational-Oriented Systems Engineering and Technology Tradeoff Analysis (ROSETTA) Environment provides a means to translate between theoretical mathematics, subject-matter expert driven analysis, and modeling and simulation, by representing a single problem using all three types of analysis and highlighting the commonalities and differences between the different representations of the problem. Since these techniques are often used in different phases of the systems engineering process, these side by side comparisons will allow feedback to be provided as to how to improve the systems engineering process for future similar studies.

For more information about ROSETTA, please contact Kelly Griendling (kelly.griendling@asdl.gatech.edu).

Research Questions

  • What is the quantitative analogy to the information gathered in qualitative systems engineering methods?
  • How can performing both a qualitative and quantitative version of the same study and comparing the results provide insight to improve future implementations of qualitative methods?
  • Where in the design process are qualitative and quantitative methods most applicable, and how can the qualitative methods be better utilized to frame the study and feed development of quantitative modeling?

The ROSETTA Methodology

Qfd.png

The subject matter expert-driven analysis will be in the form of a quality function deployment (QFD). The QFD is made from combining the seven Management and Planning Tools and is typically used in the early phases of Systems Engineering to elicit tacit knowledge from subject matter experts and use that knowledge to identify key design drivers and trades. The QFD is made of one or more hierarchical House of Quality (HOQ) tools. The HOQ that is used here maps a set of customer requirements to a set of engineering characteristics and uses the mapping to better understand the overall relationships between the requirements and between the engineering characteristics.

Modeling.png

Using modeling and simulation can provide a quantitative and more objective approach to obtaining the same information. In the ROSETTA environment, the behavior of the modeling and simulation environment in the area of interest will be captured by using the response surface method (RSM) to wrap a set of surrogate models, specifically response surface equations (RSEs), around the modeling and simulation environment. Because the RSEs are simply a polynomial regression of the model, the behavior of the modeling and simulation environment can be explored through the behavior of these simple mathematical functions. These RSEs will enable rapid (virtually instantaneous) execution of the simulation, and will allow trends across the design space to be easily understood and quantified, thus helping to determine the strengths of the relationships represented in the body of the HOQ. The RSE approach also allows for a Monte Carlo Simulation (MCS) to be rapidly conducted, and the results can be used to obtain information about the correlations of the response variables that are found in the roof and the greenhouse of the HOQ.

A more formal understanding of the information developed using the QFD method can be obtained by examining the relationships mathematically. A calculus-based viewpoint of HOQ that applies rigorous transformation formulae to describe the relationships in the HOQ is used to understand the nature of the HOQ, its implications in the overall QFD approach, and to help determine what information is required in an equivalent quantitative approach. At the same time, this mathematical approach can provide a direct translation between the information contained within the RSEs and the information elicited by the SMEs.

Publications

  • Dickerson, Charles; Mavris, Dimitri. The Relational Oriented Systems Engineering Framework Paper submitted. 6th Annual System of Systems Engineering Conference. June, 2011. Albuquerque, New Mexico.
  • Mavris, Dimitri; Griendling, Kelly. The Relational Oriented Systems Engineering Technology Tradeoff (ROSETTA) Environment Paper submitted. 6th Annual System of Systems Engineering Conference. June, 2011. Albuquerque, New Mexico.
  • Duncan, Scott; Griendling, Kelly; Mavris, Dimitri. Application of ROSETTA to the Smart Grid Paper submitted. 6th Annual System of Systems Engineering Conference. June, 2011. Albuquerque, New Mexico.