Keywords: design, geotechnical, foundation, construction, project management, expert system.

This paper describes a prototype expert system implemented to model the design process leading to choosing feasible foundations for a moderate size building. An integrated system is built providing support to geotechnical engineers during the initial design phase. The decision on the foundation type is systematically made looking at the geotechnical, construction and project management (GCPM) effects on the overall project. The system is demonstrated for a case whereby the choice of appropriate foundations is based on geotechnical, in addition to constructional and project data (cost, planning, site location and layout) prevalent in the initial foundation design stage.

Foundation systems are commonly selected during the initial design stage remaining the only design focus from then onwards. Detailed geotechnical design then proceeds where constructional and regulatory geotechnical safety requirements are fully considered. It is appropriate and greatly beneficial before embarking on any detailed design, to systematically consider the influence of for example, soil conditions, planning, various foundation-related costs, site layout and resources (constructability considerations), on foundation choices and vice versa. The interdependencies of these GCPM-type design considerations are complex to model. It's obvious that decisions made during the early stages of the construction process, including geotechnical design, will affect how a building is planned and constructed. In turn, constructability, and thus the specific choice of the construction methods to be used, will have a significant impact on the resources used and consequently the total cost of the project. In an integrated civil engineering system for foundation engineering design it is therefore required to include key data on constructability and project planning (site, cost and duration).

No reference is made in the literature to any working system that consider the foundation design process in it's totality and/or at least integrate the influence of decisions made at one of it's phases on the other phases. Surveys of Artificial intelligence (AI) research trends in geotechnics and construction[1,2,3,4] show that most models are limited to one or two fields. Reviewing current practice in the Netherlands shows that decisions on foundations are too often taken autonomously, without investigating if planning and constructability conditions can influence the choice of the foundation or vice versa.

Information technology, in particular the applied fields of Decision Support Systems (DSS) and AI, provide novel ways to construct an integrated model for selecting the most appropriate foundation system. The problem being modelled requires a combination of different modelling techniques. The system architecture adopted, and recently published[5] provide separation of processes, algorithms and knowledge in components to allow the extension of existing and addition of new knowledge and/ or functionality when required. The system comprises four main modules: INITIAL DESIGN, ALTERNATIVES, ASSESS and DISPLAY. It should be noted that it is the whole process that requires optimisation and not the individual steps, which contribute to the process. In other words, it could be beneficial in this holistic view to have sub-optimal stages in order to allow other, usually later, stages to progress with greater benefit. Traditional design objectives providing piecemeal optimisation of a project are unsatisfactory in the current system.

A construction case is incorporated in the demo whereby two foundation alternatives (a piled and a strip foundation) are assessed. The example demonstrates that the choice of a foundation system is really subject to construction aspects.

1

D.G. Toll, "Artificial intelligence applications in geotechnical engineering", in Electronic journal of Geotechnical Engineering, v.1, http://geotech.civen.okstate.edu/ejge, 1995

2

Transportation Research Board, "Use of artificial neural networks in geomechanical and pavement systems", Transportation research circular E- C012, prepared by A2K05(3) Subcommittee on neural nets and other computational intelligence-based modelling systems, Transportation Research Board, Washington, USA, 1999.

3

J.V. Cadogan, C.J. Moore, J.C. Miles, "Computational support for the design and costing of bridge foundations", Information processing in civil and structural engineering, B. Kumar (ed.), Civil-Comp Press, UK, 1996. doi:10.4203/ccp.37.4.3

4

M. Mangini, G. Varosio, E. Parker, "EP4.4 COMBI: KB-tool for foundation design", proc. ECPPM '94, Product and process modelling in the building industry", R.J. Scherer (ed.), Balkema, Rotterdam, the Netherlands, 1995.

5

J.F. van den Adel, S.H. Al-Jibouri, U.F.A. Karim, M. Mawdesley, "GROUNDSS: integrated foundation design expert system", Proceedings 8th int. conference on Computing in civil and building engineering, ASCE, 2000. This paragraph is supposed to cover the literature review

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £78 +P&P)