We have conducted series of surveys on consumer's shop-around behavior at city center retail environment (CCRE) of Fukuoka City for the purpose to create the more attractive CCRE from the viewpoint of consumer's behavior. The survey on consumer's shop-around behavior at CCRE is conducted as follows. At several sampling points (shops) chosen, samples are randomly selected from visitors at these sampling points to record their shop-around history, the sequence of shops visited, purposes done and expenditure spent there in the order of their occurrence. This is just to collect the shopping trip chain data by on-site sampling. In other words, it is simply the on-site person trip survey, which is a striking contrast to the traditional person trip survey by home-based sampling. However, as Saito, Kakoi, and Nakashima (2000) first indicated, there occurs the choice-based bias if on-site trip chain data from different sampling points are pooled to be used to estimate OD pattern. The purpose of this paper is to address this problem to formulate a new theoretical method for consistent estimation of OD pattern based on on-site trip chain data and to establish a theoretical foundation for on-site person trip survey.