This study examined weigh-in-motion (WIM) data from two states to evaluate the performance of an improved reidentification methodology that had been used to match vehicles between WIM stations. The improvement allowed the reidentification model to be trained without the use of ground truth data (i.e., true vehicle matches). The training data set was instead developed by following a three-step manual investigation of the characteristics of assumed vehicle matches between two WIM stations. The trained reidentification methodology was validated with data from Oregon, where the model was able to match identical vehicles with 70% to 90% accuracy. The reidentification was then applied to data from two WIM stations in West Virginia, where the downstream station had 6,178 vehicles with 10,427 possible matches at the upstream station. At a high confidence threshold of .98 (of 1.0), the algorithm identified 526 likely matches. Furthermore, this study examined the differential calibration of the weight sensors at each WIM location by comparing the axle weight measurements between the matched vehicles. The data from the two WIM stations in Oregon illustrated good correlation in weight measurements. However, the two WIM stations in West Virginia did not illustrate consistent relationships across all axle measurements.