Abstract:
There has been significant research in the area of automated and semi-automated regulatory compliance checking in the AEC domain over the past four decades. In order to computerise the regulatory compliance checking process, we first need to have computer representations of both the building model and the regulations. We now have Building Information Modelling (BIM) as the industry standard representation for buildings, but the challenge remains to find an efficient and practical digital representation of the regulatory knowledge. One common approach to represent regulatory knowledge for architectural and building engineering designs is to extract rules from the regulatory texts, which has proven to be quite challenging to automate. However, a popular technique used in Business Process Management (BPM) could be adapted to facilitate the representation of regulatory knowledge in the AEC domain. Additionally, the legal domain has come up with some initiatives to provide a means to share digital legislative documents and regulatory structure, e.g. Cen MetaLex, LKIF, LegalRuleML, etc., which could be useful for exchanging compliance checking data in the AEC domain. This paper outlines a practical approach to represent regulatory knowledge in terms of industry’s accepted compliant design procedures using a business process modelling technique with visual editing capabilities. The corresponding regulatory constraints, thresholds and conditional logic are treated as lookup data and rules, which could also be represented graphically. Each design procedure and its associated regulatory data would constitute the regulatory knowledge required for a particular type of design or compliance check. This paper also briefly outlines a compliance checking framework that could bring in any library of regulatory knowledge representations that are defined and maintained externally. Both the prescriptive and performance-based criteria of the New Zealand Building Code are being investigated.
