Toward resolving the problems of traditional public transport services in the case of low-density passenger flow, including high empty-loading rate, long waiting time, high operation cost, and so forth, in this paper, we propose a flexible public transport system adapted to the spatial-temporal variation of passenger flow, which takes advantage of modern information technology such as the Internet of Things and mobile Internet. This approach adopts a response mode based on requests from passengers and the automatic sensing mode of the Internet of Things to obtain dynamic information of passengers in real time and extracts the characteristics of passenger origination destination and their spatial-temporal distributions, which are used to classify the types of bus stops and formulate a parking strategy. From the aspects of vehicle scheduling, segmented scheduling and priority to unresponsive passengers in the previous vehicle are adopted, and the optimal scheduling model is established from the three aspects of total passenger income, interference loss of existing passengers, and the operation cost of the enterprise. The genetic algorithm is used to find the optimal route, and a simulation experiment based on the No. 214 bus in Guangzhou is performed for verification. The results show that the flexible public transport and scheduling method adapted to the spatial-temporal variation of passenger flow improves the revenue and expenditure ratio in comparison with the conventional flexible and traditional bus system, making it an effective way to solve the problem of public transport services in the case of low-density passenger flow.