In recent years, with the rapid development of the Internet and the accelerated pace of people's lives, takeout has emerged as a new consumption pattern. Whether it's simple fast food, high - end cuisine, or fresh produce, consumers can conveniently place orders remotely and wait for delivery riders to pick up the food from merchants and bring it to their doorsteps. As the takeout market continues to expand and the number of takeout orders surges, several issues have arisen in the traditional delivery model. During peak meal hours, order volumes often spike suddenly, leading to a shortage of delivery riders. Moreover, as customers increasingly prioritize delivery speed, riders may violate traffic rules, such as running red lights, in an attempt to meet delivery deadlines. The rapid advancement of unmanned aerial vehicle (UAV) technology has opened up a new direction for the traditional takeout delivery model—UAV-assisted takeout delivery. Taking into account China's complex urban and residential environments, the UAV-based delivery process has been optimized. After a merchant receives an order and prepares the meal, a UAV departs from the UAV distribution center to the merchant for pickup. Subsequently, the UAV transports the meal to a UAV-rider transfer point, where the rider takes over and completes the final leg of the delivery to the customer, thus creating a novel UAV-rider joint delivery model. This research formulates a model with the dual objectives of minimizing delivery costs and maximizing customer satisfaction. Delivery time is converted into cost using a penalty function. The study utilizes takeout order data from a specific area in Feixi, Hefei, Anhui Province, collected over a certain period. Given that the traditional dingo optimization algorithm is highly sensitive to the initial solution and has limited optimization capabilities, this paper proposes improvements to the initial solution and introduces resultant cross-mutation operations, giving rise to an improved dingo optimization algorithm (IDOA). The effectiveness of the proposed model and algorithm is validated through real-world numerical examples. Furthermore, the newly proposed UAV-rider joint delivery model is compared with the traditional rider-only delivery model under three different data scales. Across three computational scenarios, the new model outperforms the traditional one in terms of both the shortest delivery time and lower delivery costs. Finally, to demonstrate the superiority of IDOA, it is benchmarked against the CPLEX algorithm, the original dingo optimization algorithm, the genetic algorithm, and the genetic algorithm with an elite strategy. The results show that IDOA converges significantly faster and exhibits better optimization performance compared to the other algorithms. In terms of both the optimal value and the average value, IDOA can effectively identify the optimal solution across different data scales and complete route planning in a shorter time, confirming its superiority. In conclusion, the takeout market holds vast potential, and the novel UAV-rider joint delivery model can better satisfy customers' demands for prompt delivery while reducing delivery costs.