AHEAD:
the
Attribute
Hierarchy-based
Evaluation
of
Architectural
Designs
Method and papers (work in progress)
|
|
Karen Smiley,
Qingfeng He,
Elizabeth Kielczewski,
and
Aldo Dagnino
|
|
AHEAD is designed to complement and extend the Software Engineering Institute's
Attribute-Driven Design (ADD) method
with the increased objectivity and quantification of the
Analytic Hierarchy Process (AHP).
The AHEAD method was created by the authors to support
software systems development teams in
quantitatively analyzing a set of technology options based upon
how well they satisfy important architectural criteria.
The greater the business impact of the technology decision,
the more beneficial the AHEAD method can be.
Click the thumbnail image at left to see a
high-level diagram showing the basic steps of the AHEAD method,
see the following papers for further information, or contact the authors at
AHEAD (at) agileteams.com.
|
-
October 24, 2008:
V1.0 of AHEAD and the results of its first industrial application were
first published and presented.
|
-
Oct. 24, 2008:
"Getting AHEAD: Applying AHP for Software Technology Evaluations"
Authors: Karen Smiley, Elizabeth Kielczewski, Qingfeng He
Presented by Karen Smiley at the
20th North American Symposium on Quality Function Deployment,
Santa Fe, NM, USA. Proceedings ISBN: 1-889477-20-6
Abstract:
Evaluations of software technologies and components can be complicated,
and are easily influenced by acknowledged or latent technology biases. The Analytic
Hierarchy Process (AHP) is a natural choice for neutralizing these biases, and bringing
greater objectivity to the evaluations. This paper reports on an industrial project for
software technology evaluation which complemented the Software Engineering Institute's
(SEI) Attribute-Driven Design (ADD) technique by applying AHP for importance ratings
and for comparing the prototype implementations. We present our Attribute Hierarchy-based
Evaluation of Architectural Designs (AHEAD) method, our findings, and our analysis of the evaluation.
-
March 2009:
"Architectural Requirements Prioritization and Analysis Applied To Software Technology Evaluation"
Authors: Karen Smiley, Qingfeng He, Elizabeth Kielczewski, Aldo Dagnino
Poster accepted and submitted on Oct. 25, 2008 for the
SAC 2009 Conference -
Requirements Engineering track.
To be presented by Qingfeng He at the conference in Honolulu, HI in March 2009.
Abstract:
In this short paper, we summarize an industrial project in which we developed and applied
the Attribute Hierarchy-based Evaluation of Architectural Designs (AHEAD) method for selecting a
software technology to form the basis for the next-generation architecture of a complex commercial
software application. AHEAD leverages the Software Engineering Institute’s Attribute-Driven Design (ADD)
method and the Analytic Hierarchy Process (AHP) for evaluating software technologies that have important
architectural impact. The core activities of AHEAD include elicitation, prioritization, and analysis of
architectural requirements. The goal of these requirements activities was to establish and apply
objective criteria for selecting, prototyping, and evaluating software technology alternatives.
We found that using AHEAD brought greater objectivity to prioritization of architectural requirements
and to the technical judgments of the software technology options.
-
March 23-26, 2009:
"Getting AHEAD With DAR: How to Make Better Software Technology Selection Decisions"
Authors: Karen Smiley, Qingfeng He, Elizabeth Kielczewski
Abstract accepted on Oct. 12, 2008 for the
SEPG North America 2009 Conference.
To be presented by Karen Smiley at the conference in San Jose, CA in March 2009.
(presentation to be submitted by Dec. 12)
Abstract:
Software technology and component evaluations can be complicated,
often involving extensive criteria lists and competing stakeholder goals and priorities.
AHEAD is a rigorous multi-criteria decision method which leverages SEI’s Attribute-Driven Design (ADD)
technique and the Analytic Hierarchy Process (AHP). This session summarizes the AHEAD process,
maps it to the specific practices of the Decision Analysis and Resolution (DAR) process area of CMMI,
and describes how we applied AHEAD in an industrial case study.
|
