The work aims to propose an Open Location-Routing Problem with Simultaneous Pickup and Delivery (LOLRPSPD) model under the low-carbon background to solve the common problems of outsourcing, frequent return and exchange of goods in delivery companies and solve the problem through the Wild Horse Optimizer. Firstly, a new decoding mode was designed so that the discrete problems could be solved by continuous algorithms. Then, the initial solutions were generated through the Halton sequence to improve the nonlinear evolution probability factor TDR; The simulated binary crossover was used. Polynomial mutation operators were added. Elite preservation, and setting of consecutive failed reinitialization steps were completed. Finally, the effectiveness of the improved algorithm was verified through the comparison results between Improved Wild Horse Optimizer (IWHO), Wild Horse Optimizer (WHO), Simulated Annealing (SA), Particle Swarm Optimization (PSO), and genetic algorithm (GA). The experimental results showed that IWHO in this paper had better optimization ability than normal WHO in terms of large and medium-sized examples, and had a good advantage in the processing of small examples whiling ensuring accuracy. The proposed LOLRPSD model is reasonable, and the Improved Wild Horse Optimizer has better searching ability for LRP problems.