In order to explore all the possibilities of civil cargo transport aircraft for loading military equipment, improve the ability to respond quickly to the needs of operations, and provide the necessary military equipment and various types of protective materials for operational success, weight and balance problems of civil-to-military transport aircraft are studied. To determine the loading position and orientation of military equipment in the cargo hold of an aircraft, the civil wide-body freight aircraft B747-400F modified model is selected as the research object. An optimization model that integrates a two-dimensional geometric model with the load balance of transport aircraft is established by utilizing a two-stage decomposition strategy based on the two-dimensional cutting theory in consideration of the cargo hold of the transport aircraft as a rectangular plate, and military equipment as multiple rectangular items with varying lengths and widths. In the two-dimensional geometric position constraint model, constraints are considered for the non-overlapping positions of military equipment, maintaining a certain distance, not exceeding the cargo hold boundaries, and the possibility of orthogonal rotation. In the load balance optimization model, the objective function is to minimize the center of gravity (CG) deviation and maximize the payload in consideration of constraints such as the transport aircraft’s weight, payload, CG, cargo hold cumulative load, linear load, and area load. A Benders decomposition algorithm is designed to decompose the load balance optimization model with complex issues containment into a master problem in position allocation and a subproblem for weight balance checking. Gurobi is used to solve two example scenarios for model verification. The results indicate that the established model and the algorithm can quickly effectively determine the loading position and the orientation of military equipment under a reasonable CG, improving the payload, and providing references and insights for intelligent loading.