A knowledge-based cost estimating system for manufactured residential housing

Abstract

Cost estimation is a critical part of the manufacturing process. This is more so in the case of manufactured or "prefabricated" housing. A cost estimation approach for obtaining a quick, accurate estimate has been developed in this project. Prior to the development, literature in the areas of expert systems, construction cost estimation and artificial intelligence were read and the most relevant papers have been reviewed in the report. The estimation approach has been developed based primarily on "knowledge" derived from a firm manufacturing residential housing. It is formed based on relationships between the various sections of the building and uses formulae developed based on this "knowledge" and standard cost estimation relationships for calculating a total cost estimate. The formulae use the primary cost drivers for each section of the construction work to arrive at the estimate.

The whole process has been delivered on an expert system platform. The expert system has been developed using a windows-based expert system shell, Kappa-PC™ produced by Intellicorp™. The report contains, other than the description of the approach mentioned above, the description of the expert system which was developed, and the development methodology. A case study is also given for the purpose of a better understanding of the use of the expert system and to prove the effectiveness of the approach and formulae developed.

The complete process has been shown to have an accuracy of about 5 % based on the difference between the values calculated by a human expert and the values obtained using the formulae and procedure developed for the same input.