Load and Resistance Factor Design (LRFD) for Deep Foundations (2004)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

71 CHAPTER 4 CONCLUSIONS AND SUGGESTED RESEARCH 4.1 CONCLUSIONS The data and analyses presented in this report lead to the following major conclusions: 1. The compilation of large databases allows for the quan- titative assessment of pile capacity evaluation methods during both design and construction. In addition, data- bases (combined with the application of mechanics principles) allow the determination of the controlling parameters of capacity evaluation methods that require calibration. Databases are essential therefore for the examination of any design methodology and hence enable the testing of the factors of safety used by WSD (i.e., their validity and effect on costs) and the devel- opment of other methodologies, such as LRFD. 2. LRFD facilitates a design methodology that is more suitable for geotechnical applications than WSD. The ability to determine design factors while quantifying the significance of their outcome is a powerful tool in engineering. The development of load and resistance factors utilizing reliability-based calibration and data- bases is a major necessary step toward objective quan- tification of the LRFD parameters. More so, it allows a meaningful utilization of the LRFD principles (in con- trast to parameter fitting to WSD) and sets a base for future developments that will further rationalize design and lead to more economical construction. 3. The findings presented are the first of their kind in the development of resistance factors for LRFD design methodologies in geotechnical engineering. A review of the recommended resistance factors must be compatible with the fundamental principle of the methodology, i.e. engineering design with a consistent level of reliability. Existing LRFD codes worldwide were developed, by and large, to be compatible with previous WSD parame- ters based on different concepts. While radical changes cannot be expected in construction practices, a shift in both directions (more and less conservative depending on the specifics) should be expected and accepted when adopting a true LRFD design. 4.2 SUGGESTED RESEARCH— KNOWLEDGE-BASED DESIGNS 4.2.1 Statement of Problem Variability in the parameters used in the design, site con- ditions, construction quality, and previous experience are all important factors. The present study bypassed some of the difficulties by calibrating specific design methods and corre- lations as a unit. A more complete design process based on LRFD can recognize the contribution of different factors— such as subsurface variability, site-specific technology, and previous experience—as well as amount and type of testing during construction. A framework for such an approach is presented here; further development, however, will require additional databases, e.g., for the correlation between soil parameters and field tests, as well as subjective judgments. 4.2.2 Framework for LRFD Design for Deep Foundations In order to fully exploit the potential of the LRFD method- ology for geotechnical purposes the aforementioned issues must be addressed. Many of the affecting factors are in fact being considered in the design (e.g., previous experience) but need a framework to allow future progress. A proposed solu- tion is to establish knowledge-based factors for both the design and the construction (independent) capacity evaluation methods. These factors can be accounted for by a modifying constant ξ to be multiplied by the resistance factor. ξφ R > γL (40) Where: ξ = ξ1 ξ2 ξ3 ξ4 ≤ ξlimit ξ1 = factor adjusting for the variability of site conditions ξ2 = factor adjusting for the quality of soil parameter estimates ξ3 = factor adjusting for construction quality control ξ4 = factor adjusting for previous site or construction experience

ξlimit = an upper limit on the factor that will be determined from computing the components of ξ (and judg- ment). The limit should have some real value larger than 1.0, such as 1.10. A short description of the knowledge-based factors follows. ξ1 applies to the spatial variation of soil properties, strati- fication across a site, and the extent to which that variation has been categorized by the subsurface investigation pro- gram. The factor is relevant for deep foundations capacity evaluation during both design and construction. Low values mean that the site is more erratic than normal or that little exploration and testing has been done. Average values reflect normally variable soil conditions adequately investigated. High values mean that the site is more uniform than normal and that an extensive program of boring and testing has been conducted. The extent of exploration can be evaluated via number of borings per substructure unit. ξ2 applies only to deep foundations capacity methods employing calculations based on soil parameters. This factor accounts for the manner in which soil parameters are estimated from field and laboratory test data and the exactness of those estimates. Low values mean that the correlation between soil parameters and the measurements they are based upon is poor (e.g., when estimating undrained shear strength of soft to 72 medium clay from SPT values), and thus the confidence in the accuracy of the soil parameter values is low. Average values reflect normally variable soil conditions adequately inves- tigated. High values mean that the correlation is good (e.g., when parameters are estimated based on laboratory test of undisturbed samples or direct in situ testing like a field vane test), and thus the confidence in parameter values is high. ξ3 applies to the extent of measures taken to control con- struction quality and testing the integrity and capacity of the constructed deep foundations. Low values mean that few measures are to control construction quality; and no static, dynamic, or integrity testing results are available. High val- ues mean that extensive measures are taken to control con- struction quality, and multiple pile testing results are avail- able (e.g., integrity and capacity of drilled shafts, static and dynamic tests of driven piles). The high quality control also relates to the number of tested deep foundations as a ratio of the number of piles installed per substructure. ξ4 is to be used during the design to account for previous experience accumulated either on a specific construction site or from a specific construction technology. Low values are used if no previous experience is known at the site and a new unfamiliar construction technology is used. High values mean that previous deep foundation testing results similar to the one designed (type and installation) are available.