Supporting Design Rationale Retrieval for Design Knowledge Re-Use

Design rationale records the issues addressed, the options considered, and the reasons for making the specific decisions during a design process. However, it is often difficult for a designer to retrieve a particular piece of rationale from a large dataset, which makes the computer-aided retrieval a necessity. Current retrieval methods focus mainly on either the classification of rationale or on keyword based searches of records. The pitfalls of these methods are obvious. First, a classification scheme imposes a large burden on experts. Second, key-word based searches require the designers to understand, at least to some extent, what they are seeking. This ongoing research aims to develop methods and tools for designers to facilitate the re-use of design rationale captured via a software tool currently used in industry. Our first step is to improve the key-word based search by providing better human-computer interaction so that designers can find some useful information even if they cannot explicitly express their information needs. Specifically, the improved retrieval is implemented by a novel combination of three methods, namely suggesting potential key-words to designers, quantifying the relevance of retrieved information, and recommending relevant information based on the dependencies. A software prototype has been developed to demonstrate the proposed approach, as well as to research into how it can be integrated into the design rationale capture tool. Preliminary evaluation shows the prototype can provide good supports to designers especially when they do not know exactly what to search for. Some problems encountered during this work are also identified in the evaluation, leading to a discussion of our future work.