Poirot:TraceMaker utilizes a probabilistic approach to dynamically retrieve traceability links between artifacts of the software development process.  The probabilistic inference model at the core of the tool calculates the probability of a link between a query and document based on the commonality of terms between the query and document, the general distribution of those terms in the dataset, and also through incorporating knowledge about the clustering and hierarchical structuring of artifacts.

Poirot is a web-based tool designed to provide efficient traceability support for all stakeholders within an organization  It provides a streamlined and simple interface to a relatively complex underlying inference engine.. External artifacts are processed either through their XMI representation or through a customized interface.  Poirot supports a wide variety of  trace queries displaying results in both a textual and visual format.  This demo will briefly explain the theory of the underlying probabilistic model and demonstrate its support for requirements validation and impact analysis in both a traditional and agile development environment.

Related publications:

• Jane Cleland-Huang, Raffaella Settimi, Oussama BenKhadra, Eugenia Berezhan, Selvia Christina, “Goal Centric Traceability for Managing Non-Functional Requirements”, Accepted at International Conference on Software Engineering, St Louis, USA, May 2005.

• Jane Cleland-Huang, Grant Zemont, and Wiktor Lukasik, “Heterogeneous Solutions for Improving the ROI of Requirements Traceability”, IEEE International Requirements Engineering Conference, Kyoto, Japan, Sept. 8-10, 2004.

• Raffaella Settimi, Jane Cleland-Huang, Oussama BenKhadra, Jigar Mody, Wiktor Lukasik, and Chris DePalma, “Supporting Change in Evolving Software Systems through Dynamic Traces to UML”, IEEE International Workshop on Principles of Software Evolution, Kyoto, Japan, Sept 6-7, 2004

Download an overview of the Poirot:Tracemaker tool

An investigation into the interdependency between requirements and software architecture.

IFM is an ROI-informed approach to software development, in which software is developed and delivered in carefully prioritized chunks of customer valued functionality known as Minimum Marketable Features or MMFs. IFM integrates traditional software engineering activities with financially informed project management strategies in order to optimize critical metrics such as NPV and ROI.

View a recent review on Software by Numbers

EBT is an event based approach to traceability, in which Software Engineering work products are dynamically linked using the publish-subscribe paradigm in place of a traditional traceability matrix. EBT supports a high-degree of automation for tasks such as impact analysis and regression testing, and solves many problems normally related to the long-term maintenance of traceability links.