The complex problems of predicting adhesive bond strength for both adhesive and cohesive defects have been studied using an ultrasonic, experimental test bed system. This experimental test bed incorporates the ultrasonic and computer equipment necessary to acquire and process data from various types of adhesively bonded test specimens. The computer hardware and software have been developed to allow the design of reliable pattern recognition algorithms for the prediction of adhesive bond strength. Two different types of adhesive bonded structures were studied. First, the problem of inspecting the adhesive bond joint in an aluminum to aluminum step-lap specimen to predict the bond strength that could be affected by adhesive or cohesive defects was studied. A set of 164 bond specimens was used to design an algorithm that is 91% reliable for separating the specimens into a good class or a weak class. A Fisher Linear Discriminant function was selected by the test bed system as the optimal pattern recognition routine for the classification problem. The second structure studied is the honeycomb configuration. Specimens were acquired that contained many of the typical adhesive defects common to honeycomb structures. A feasibility study was conducted to determine the test bed's potential for solving honeycomb inspection problems.