Computer aided design document retrival by data mining technique

The advancement and widespread application of information technologies have been generating vast amount of electronic engineering documents. Among them, the computer aided design CAD documents may be the top ranked in terms of its quantity. The CAD documents are generated by engineers or architects while performing engineering tasks e.g conceptual planning, basic design, detailed design, and construction supervision. The quantity of CAD documents being generated usually depends on the type and size of the project. A typical five-floor residential building may require less than 100 CAD drawings. However, a mass transportation project may generate more than 200,000 CAD documents.

Due to massive growth of CAD documents, construction organizations are facing an increasing management costs needed both for storage and retrieval of the electrical CAD documents. The importance of the CAD document can be viewed from three aspects:

1. Provides an effective communication medium to illustrate the design concept of an engineering product, so that engineers and architects can “visualize” their ideas.

2. It is a legal document that provides a basis for performing, management, and closure of a contract.

3. Provides a useful library for engineers and architects to reuse previous design models in order to accomplish their design efficiently.

In construction practice, the cost items not included in the CAD documents are considered extra work that needs to be tackled with change orders. Moreover, when integrated with construction schedule, the CAD documents provide further help to the construction planner for progress control and dynamic resource allocation.

The computer aided design CAD document provides an effective communication medium, a legal contract document, and a reusable design case for a construction project. Due to technological advancements in CAD industry, the volume of CAD documents has been increased dramatically in the database of construction organizations. Traditional retrieval methods relied on textual naming and indexing schemes that require the designers to memorize in details the meta-information used to characterize the drawings. Such approaches easily overwhelmed the user’s memory capability and thus caused low re-usability of CAD documents. Content based text mining technique can be adopted to extract the textual content of a CAD document into a characteristic document, which can be retrieved with similarity matching using a Vector Space Model, so that the automated and expedited retrievals of CAD documents from vast CAD databases becomes possible. A prototype system, namely content-based CAD document Retrieval System, is developed to implement the proposed method. After preliminary testing with a CAD database and a public engineering drawing database, the proposed content-based CAD document retrieval system is proven to retrieve all relevant CAD documents with relatively high precision.

References:

Department of Construction Management, Chung Hua University, Hsinchu 300, Taiwan, ROC