[Purposes] Addressing the contradiction between declining passenger traffic and rising logistics demand in rural areas, this study proposes an innovative integrated passenger and freight transport model that incorporates crowdsourced vehicles. The model aims to optimize resource allocation, enhance logistics efficiency, and alleviate the "last-mile" delivery problem in rural regions. [Methods] A vehicle scheduling model is formulated to minimize total costs, including truck transportation costs, cargo penalty costs, passenger delay costs, and crowdsourced vehicle detour costs. The model is subject to constraints such as time windows, truck capacity, and simultaneous pickup and delivery. An improved tabu search algorithm is designed to solve the model, which enhances computational efficiency through dynamic tabu length adjustment and an adaptive penalty mechanism. [Findings] An empirical analysis in Linli County, Hunan Province, demonstrates that the proposed model effectively reduces transportation costs while significantly improving service coverage and cargo turnover efficiency. Sensitivity analysis shows that in areas with scattered demand, crowdsourced vehicles can reduce unit freight cost by 23.7%. Moreover, when the participation rate of crowdsourced vehicles exceeds 40%, the overall delivery timeliness improves by 34.2%. [Conclusions] The study confirms that integrating a crowdsourcing mechanism into the passenger and freight synergy can optimize logistics resource allocation, mitigate last-mile delivery challenges in rural areas, and provide a theoretical foundation and practical path for the sustainable development of urban-rural transport.