Dr. Joel Sercel spoke on “The Systems Engineering Factory: Examining Depth, Disruption, and Scope to Resolve the Paradox of Today’s Acquisition Failures”. Dr. Sercel is the Technical Director of Systems Engineering Training for California Institute of Technology, Industrial Relations Center, 383 South Hill Street, Pasadena, California http://www.irc.caltech.edu/systemsengineeringcenter.
Sercel defined D2S (Depth, Disruption and Scope) as a 3-axis model that can be used to evaluate any project or program. A project’s position in D2S space determines the fundamental values, process, management approaches, and corporate culture needed to ensure the most successful possible program execution. A cursory understanding of the D2S model can be had by considering what each of the individual scales in the model tells us. For example, the Depth scale relates to the extent of Discovery (uncharted territory) required to complete a program, and hence the degree to which erudition and respect for scientific peer review is needed.
Degree of Disruption of Key Ideas embedded in the project is an indicator of the extent to which the project execution team should be Risk Embracing or Risk Avoiding to maximize performance. Program Scope determines the extent to which formal Process Discipline, such as the systems engineering methods promulgated by INCOSE are needed for the project. For very small, simple projects Sercel suggest that ad hoc processes are often most effective, while for very large programs such as many of today's defense systems developments, formal process discipline and methods become keys to success.
What Doesn’t Work (For Large Programs)? Sercel noted the following practices in current use that are not often effective: ‘Armies of System Engineering Technical Assistants (SETAs)’ who are non-government employees brought in to conduct system engineering activities often independently; ‘Death by MURDER Board’ where a presentation or document is vetted by a committee until no one can find any objections; ‘Requirements By Telecon’, where candidate requirements are massaged oneby-one in a large group—often including representatives of multiple disciplines, some not necessary —until a consensus is achieved.
Sercel contrasted those practices with what he calls the Systems Engineering Factory. For large programs the SE factory consists of: Facilities, Tools, Teams, Processes, Procedures, Standard Products and Business Rhythms, designed to be compatible with the D2S characteristics of the project and based upon systems engineering best practices. Joel emphasized the ‘Business Rhythms’ define the synchronization of the processes used during the development. In SE Factory implementation, best practices include: Developing a complete document/product tree, Designing the required documents, Assigning the leaders, Estimating the effort required -- based on past experience (not wishful thinking), Writing needed documents, and Measuring progress. In a sense, applying systems-engineering techniques to the activities of systems engineering, thus designing a system to do systems engineering.
Presentation Evaluation: Sercel’s talk was one of the most important discussions we have had at INCOSE-LA because it dealt with a major conflict within the systems engineering community.
On one side of the conflict is the large number of requirements that have to be allocated down through the several layers of the system architecture, analyzed through trade studies, validated, and verified among other tasks at each level. This need often leads to costly, time-consuming, and labor-intensive processes to coordinate the results. In practice, this can lead to cost overruns, schedule slips, and technical failures.
On the other side of the conflict is a product-oriented view that tends to disregard process and culture and therefore the lessons learned that have created modern systems engineering.
Sercel’s method, which he calls the SE Factory as briefly described above, is offered as a prescription for the resolution of the conflict on large programs. The SE Factory concept is a worthy hypothesis and deserves the validation and verification through actual practice on a wide range of projects. Sercel also made in clear in his presentation that "one size does not fit all in systems engineering" and that the systems engineering factory and INCOSE methods are not appropriate to all programs. For efforts with high scientific depth, or for small, informal projects, INCOSE methods must be tailored down to reflect the cultural needs of the effort. The D2S model provides a way to guide this tailoring and resolve the schism facing the systems engineering community today.
Sercel has indicated that his full tutorial expands upon the brief overview presented in the meeting. The INCOSE-LA chapter is checking into availability and pricing of having a D2S model and SE Factory tutorial in 2008. Stay tuned. *** JPL and Caltech provided the INCOSE-LA chapter with the capability to video record, videostream and webcast this meeting. The INCOSE-LA chapter profusely thanks JPL and Caltech for providing these services. The streaming video worked beautifully for the audience at the Aerospace remote site. The Boeing remote sites lost the video 1/3 of the way through the meeting and were not able to re-establish the video, but were able to continue with the audio webcast. Even with these difficulties Dr. Sercel’s presentation was as effectively delivered to the remote sites as at the face-to-face location. The July 22 meeting served nearly 70 participants who were distributed across the following locations: JPL and our remote webcast receive sites at The Aerospace Corporation, Boeing Huntington Beach, Boeing Anaheim and the Antelope Valley College. Contributing authors: Shirley Tseng, Dr. Scott Jackson, Dr. Joel Sercel, Dick Emerson |