Abstract: The traditional fresh food inventory vehicle scheduling problem usually adopts a non-interrupted loading/unloading mode, which can easily lead to the problem of unloading non-required goods and the failure to unload required goods in a timely manner. To accelerate the cross-warehouse transportation of fresh products and reduce the loss of freshness during the transshipment process in cross-docking systems, this paper proposes an interrupt mode for inbound trucks, i.e., the unloading process of inbound trucks can be interrupted, meanwhile, establishes a truck scheduling model of fresh produce with the goal of minimizing the total freshness deterioration in cross-docking systems. According to the interrupt mode for inbound trucks, a three-layer hybrid chromosome coding scheme is designed by taking the inbound truck serial number, fresh produce types, and batch unloading quantity as genes. Thereby, an adaptive genetic algorithm based on batch unloading quantity is constructed to solve the problem. Numerical experimental results show that, compared with the traditional non-interrupt mode, the interrupt mode of the inbound trucks can effectively reduce the total deterioration by an average of 13.90% during the cross-docking transshipment process of fresh produce, and this advantage will become more significant with the decrease of truck changeover time, transshipment time, batch unloading/loading quantity, and the proportion of high deterioration rate of fresh products.