In order to promote the development of low-altitude economy and improve the economic efficiency of enterprises and the efficiency of the “last-mile” delivery, the multi-trip truck-drone routing problem with profits and time windows is proposed. Firstly, a basic mixed-integer programming model (MIP) of the problem is established, considering the maximization of revenue, and the slack of the basic model is reduced by incorporating multiple valid inequalities. Then, an effective hybrid heuristic algorithm is proposed to solve the problem, and considering that the truck-drone paths with time windows characteristics are more complex and the feasibility checking is time-consuming, a Segment-based efficient evaluation method is designed to accelerate the feasibility checking of the paths in order to improve the solving efficiency of the algorithm. The experimental results show that the valid inequalities improve the speed of solving the model by 44 % for Gurobi; The proposed algorithm outperforms Gurobi and the two types of heuristic comparison algorithms for solving efficiency and solution quality in different sizes of instances, and exhibits good stability; Moreover, the algorithm's solving time can be reduced by 95 % by embedding the Segment-based efficient evaluation method.