Dynamic pile driving formulas have been available for the field prediction of the static bearing capacity of pile foundations for well over 180 years. On account of the immense number of different formulas that have been amassed during this time frame, a review of published literature was performed to identify the most common dynamic pile driving formulas utilized in the United States and their documented reliability. The results of this review indicated that no one dynamic pile driving formula is consistently better than all the rest; however, the Hiley, Janbu, Pacific Coast Uniform Building Code (PCUBC), and Gates formulas were shown to provide, on average, the best predictions of ultimate pile capacity. In contrast, the Engineering News Record (ENR) formula, which has been probably the most widely used dynamic formula within the United States, was shown to be among the worst predictors of pile capacity.

For well over 100 years, the Working Stress Design (WSD) approach has been the traditional basis for geotechnical design with regard to settlements or failure conditions. However, considerable effort has been put forth over the past couple of decades in relation to the adoption of the Load and Resistance Factor Design (LRFD) approach into geotechnical design. With the goal of producing engineered designs with consistent levels of reliability, the Federal Highway Administration (FHWA) issued a policy memorandum on June 28, 2000, requiring all new bridges initiated after October 1, 2007, to be designed according to the LRFD approach. Likewise, regionally calibrated LRFD resistance factors have been permitted by the American Association of State Highway Officials (AASHTO) to improve the economy of bridge foundation elements. Thus, the bulk of this study focused on the development of regionally calibrated LRFD resistance factors for the construction control of driven pile foundations via a suitable dynamic pile driving formula.

Using data from pile load tests performed in the State of Iowa, which was analyzed for reliability and placed in a newly designed relational database management system termed PILOT-IA, the efficiency of seven dynamic pile driving formulas (i.e., the Gates, FHWA Gates, ENR, Iowa Department of Transportation (DOT) Modified ENR, Janbu, PCUBC, and Washington DOT (WSDOT) formulas) was investigated. In addition to verifying the poor performance of the ENR formula, it was demonstrated that the efficiency of the Iowa DOT Modified ENR formula, which is presently specified in the Iowa DOT's Standard Specifications for Highway and Bridge Construction manual, is sufficient to allow for its recommended use with steel H-shaped and timber pile foundations driven in any soil type; these two driven pile foundation types were found to be the most commonly used in Iowa via the results of both a state- and county-level survey. More specifically, LRFD resistance factors were calibrated and verified on a pile and soil type basis for the Iowa DOT Modified ENR formula using the first-order, second-moment (FOSM) reliability approach and the findings obtained from nine full-scale field load tests performed throughout the State of Iowa on steel H-shaped piles. For a target probability of failure of 1%, LRFD resistance factors of 0.49, 0.62, and 0.50 have been recommended for use with steel H-shaped piles driven in sand, clay, and mixed soil profiles, respectively, with a factor of 0.35 having been cautiously recommended for use with timber piles driven in any soil type.

Finally, a displacement-based signal matching technique was recommended for use with PDA measured data to arrive at prediction correlations for soil quake values, Smith damping factors, and the degree-of-degradation of such parameters with respect to pile penetration depth. Although an insufficient amount of data was analyzed to even begin to develop such correlations, the accuracy, uniqueness, and theoretical basis of the displacement-based signal matching approach over the more commonly employed Case Pile Wave Analysis Program (CAPWAP) approach was demonstrated. Provided the future establishment of such prediction correlations from the increased use of this proposed technique, it was suggested that a one-dimensional pile-soil model could be used in conjunction with a dynamic pile driving formula to design driven pile foundations.