This thesis presents a novel method for the correlation of FEM results to experimental test results known as the "Load updating method." Specifically, the load updating method uses the math model from the FEM and the strains measured from experimental or flight test data as inputs and then predicts the loads in the FEM which would result in strains that would correlate best to the measured strains in the least squared sense. In this research, the load updating method is applied to the analysis of a complex frame structure whose validation is challenging due to the complex nature of its structural behavior, load distributions, and error derived from residual strains. A FEM created for this structure is used to generate strain data for thirty-two different load cases. These same thirty-two load cases are replicated in an experimental setup consisting of the frame, supporting structure, and thirty actuators which are used to load the frame according to the specifications for each of the thirty-two load conditions. A force-strain matrix is created from the math model in NASTRAN using unit loads which are separately applied to each load point in order to extract strain results for each of the locations of the seventy-four strain gages. The strain data from the structural test and the force-strain matrix is then input into a Matlab code which is created to perform the load updating method. This algorithm delivers a set of coefficients which in turn gives the updated loads. These loads are applied to the FEM and the strain values extracted for correlation to the strains from test data. It is found that the load updating method applied to this structure produces strains which correlate well to the experimental strain data. Although the loads found using the load updating method do not perfectly match those which are applied during the test, this error is primarily attributed to residual strains within the structure. In summary, the load updating method provides a way to predict loads which, when applied to the FEM, would result in strains that correlate best to the experimental strains. Ultimately, this method could prove especially useful for predicting loads in experimental and flight test structures and could aid greatly in the Federal Aviation Administration (FAA) certification process. / Master of Science / The research presented in this thesis provides a new way for correlating data obtained during structural testing with results obtained from computer analysis known as the finite element method (FEM). During the process of certifying an aircraft structure with the FAA, it is important to be able to demonstrate that the results obtained for a given structure with a computer model matches the results produced by a real world experiment within a reasonable tolerance. Traditionally, differences between these two results have been accounted for by adjusting the model within the computer until its results match those from the test. However, in this research the loads which are applied on the computer model are changed instead until loads are found which produce results in the computer models that match those from testing. This method, known as the load updating method, therefore provides a way to predict loads on a structure where the loads are unknown such as a flight test article. Here, the ability of the load updating method to predict loads on a complex three dimensional frame structure is explored and the accuracy of the results studied by comparing the results to those from a structural test whose loads are known. It was found that the load updating method does indeed predict unknown loads to a reasonable accuracy and could aid future design efforts immensely.