As early as 1993 AGS identified the need for a detailed guidance document on laboratory testing, says Richard Thomas, chairman of the AGS working party set up to draft the new guide 'Selection of Geotechnical Soil Laboratory Testing'. British Standards, mainly BS 1377 (1990), provide very detailed procedures for carrying out tests, but give no advice on the scheduling laboratory tests, and geotechnical textbooks on ground investigation typically pay little regard to laboratory testing.
There is little advice on the type and quantity of testing appropriate for the classification of soils or for the determination of design parameters. Whereas experienced geotechnical engineers have in the course of their careers developed ideas on the type and quantity of tests to schedule for particular problems, there has been very little written guidance, and nothing that brings the information together.
For the guide to be of real benefit, says Thomas, the AGS working party was first faced with the problem of establishing current best practice in the UK, particularly with respect to the quantity of testing carried out for a particular application. The general view was that engineers usually spent the available budget, but how that budget sum had been determined was often unclear or unknown.
While it was recognised that all projects are different it was felt that it should be possible to give some guidance as to what was normally con- sidered to be an acceptable quan- tity of testing.
The first step was to see if the individual mem- bers of the working party could agree on what constit- tuted a suitable scope for testing. A hypothetical project was drawn up and each member of the working party then prepared their own testing schedule. The result - surprisingly for some - was a high degree of agreement.
The working party then approached its membership to collect data on testing expenditure for various types of project. The results were disappointing, says Thomas, because few organ- isations maintained adequate records to enable the information to be collated easily. However one major employer, two contractors and a leading consultant were able to provide meaningful information on construction sectors.
This identified geotechnical soils laboratory testing costs for routine work in the building and water industries to be very similar and mainly comprise simple classification, chemical and undrained strength tests. For highway schemes expenditure on testing is greater, because of the increased need for compaction related and effective stress tests.
On the basis of these findings the AGS guide recommends that while each project must be considered individually a mimimum allowance of 12% to 15% of the total investigation cost (as paid by the contractor but excluding contamination testing) be made for geotechnical testing on the majority of relatively straightforward projects. However a minimum of 25% of investigation costs may be needed where the project includes earth- works, cuttings, deep excavations or retaining walls, unless the emphasis is to be placed on insitu testing or comparable experience.
Richard Thomas is director of Foundation & Exploration Services