TR157: CoMem: Design Knowledge Reuse from a Corporate Memory

The objective of this research is to improve and support the process of design knowledge reuse in the architecture, engineering, and construction industry. Whereas internal knowledge reuse from one’s personal memory is effective, external knowledge reuse from an external archive often fails. Ethnographic observations from this research identified three steps of internal knowledge reuse:

• Finding a reusable item

• Exploring this item’s project context which leads to understanding

• Exploring this item’s evolution history which leads to understanding

The external reuse process is designed so that it matches the internal reuse process. The hypothesis is that if the designer’s interaction with the external repository enables him/her to find and understand items of design knowledge then the process of reuse will be improved.

A CoMem (Corporate Memory) prototype is presented. It consists of three modules to support the three steps: (i) find, (ii) explore project context, and (iii) explore evolution history.

The CoMem Overview supports the finding of reusable knowledge in external repositories. It is implemented as a treemap where each item is color-coded by its relevance to the user’s design task. The relevance measure is generated using traditional as well as innovative semantic information retrieval techniques.

For any given item, the CoMem Project Context Explorer identifies related items in the corporate memory, and visualizes these related items to help the user better understand why the item in question was designed the way it was.

The CoMem Evolution History Explorer presents the versions of the selected item, and the team interactions and rationale driving this evolution. It draws from the effectiveness of storytelling, and explores how to visualize version histories.

A usability evaluation of CoMem is performed using formal user testing. The results show that CoMem offers greater support for finding and understanding than traditional tools. Reuse using CoMem is consistently rated more effective by test participants. This supports the hypothesis that finding and understanding lead to more effective reuse. This research makes important contributions by formalizing the reuse process, developing an innovative tool to support that process, and building a framework to study and assess such tools.