Design by Numbers: Quantifying the Qualitative

By:

Design is an open-ended process to create innovative systems or tools that address the needs of stakeholders subject to legal, organizational, social, and aesthetic constraints and objectives. Communication is critical in order to bridge the gap between different perspectives and express the design as the resolution of competing interests. Courses that teach design are challenged to provide a context in which the open-ended nature of the design process can be experienced, while still imparting clear direction, structured evaluation and principled guidance for students. Evaluation of a design is the essential challenge. How can one measure a design to determine whether it is good in comparison to other designs? What criteria can be used to measure both the design and the design process? When is a design successful?

In this paper, we examine answers to such questions based on the experiences of a final year design project course, in which software engineering students design and deploy a complete software system on behalf of a customer over an eight month period. Students are responsible for creating their own teams, finding their own customers and defining their own projects including the criteria for success. Given the wide range of projects that result, a systematic approach to evaluation that is fair and consistent is challenging. Clear communication of expectations and a consistent approach to evaluation is based on a continuous process of negotiation of objective, quantifiable measures. In this context, the final design that emerges from successful projects is a clear, precise expression of the qualitative requirements and objectives for the project in terms of quantifiable and verifiable measures against an evolving set of precise deliverables. Design, not as a chaotic process, but rather as chaos processed.

Stream: Teaching and Learning

Paper: , Design by Numbers

Liam Peyton, PhD, P.Eng., is the principal investigator for the Intelligent Data Warehouse laboratory at the University of Ottawa. He teaches courses in software engineering design and electronic commerce. He has 10 years experience as an industry consultant specializing in innovation, knowledge transfer, and mentoring on projects that address the requirements of specific industries as they adopt new technology to automate and manage business processes He has degrees fromAalborg Universitet (Ph.D. 1996), Stanford University (M.Sc. 1989), and McGill Unviersity (B.Sc. 1984).