Feasibility of Representing Selected Elements of The 1985 Building Regulations in Prolog or other Rule-Based Form

Keevil, Philip (1998). Feasibility of Representing Selected Elements of The 1985 Building Regulations in Prolog or other Rule-Based Form. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000fea9


This dissertation examines the proposition that there is a consistent and repeatable internal structure within regulations. This is demonstrated to be the case for regulations which make up The Building Regulations 985, which is used to illustrate many of the arguments discussed. The outcome of the research is the representation of a series of example regulations to reveal the nature of their internal structure. The form that the internal structure takes is embodied in a series of linked data structures known as 'frames' using the Prolog computer language.
Evidence for the existence of an internal structure is based upon data obtained from analysis of one year's Appeals and Determinations in a report taken from a series published from time to time by the Institute of Building Control.
Theories about the nature of prescriptive rules are discussed to introduce the subject of regulations which are described in the light of relevant legal theories about rules in general. To clarify the issues involved a systems model is proposed to provide an overview of the building regulations process. The history of regulatory control of building construction in this country is described as a prelude to considering the nature of disputes and the type of information they can yield to support the thesis. Research into the application of Artificial Intelligence techniques for processing legal statements is reviewed to establish how such experiments relate to the concept of internal structure in regulations. The appraisal identifies a number of representational problems that restrict presentation of the internal structure. The dissertation is supported by worked examples of regulations that have been analysed and by output from a sample analysis session.

Viewing alternatives

Download history


Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions