The SEI's work in process management supports data-centric approaches to advance the discipline of software engineering. When practitioners believe that the development process is inherently unpredictable and unmeasureable, defects are inevitable, and quality costs more, they develop solutions to the wrong problems; solutions that don't advance software engineering as a science.

The SEI is working to change the standards for how software is built by focusing on the data that underpins every research effort and should guide every management and development decision. We plan to raise the bar on software engineering practice, and hold ourselves to the same standard as we develop and evaluate models, methods, and practices.

To support this goal we work in these areas:

• Providing guidance for measurement frameworks compatible with multiple development methods
• Working with government and industry on the collection and analysis of fine-grained, high quality data for analysis
• Conducting research in the application of advanced analysis techniques, primarily to model and forecast development and sustainment costs and outcomes
• Developing performance-focused, measurement-driven models of best practices
• Guiding the development of agile, integrated solutions built by others through empirical evaluation