Logistics systems in the Serbian Armed Forces are built in order to ensure and maintain combat readiness. During combat actions the structure of logistics forces, equipment and resources is organized in order to ensure success in combats and operations. Progress in information security and transport technology makes it possible for a soldier to switch mass for speed and to be sure that everything will work well. The spectrum of a full support means the support to a soldier from the supply source to the place where it will be needed. In order to obtain appropriate systems for logistics support, the systems which meet requirements and which are adjusted in accordance with environment changes and new requests are created, notably models based on the operational research methods. The key point in the process of transport management in the Serbian Armed Forces is a decision making process. On a daily basis, the units of transport support obtain a large number of requests from other units of the Serbian Armed Forces demanding the transport of different types of load to different destinations. Each transport request is characterized with a number of attributes such as: type of goods, quantity (weight and volume), places of loading and unloading, expected time for loading and/or unloading and distance to which goods have to be transported. This paper shows a neuro-fuzzy model as a support to the decision making process. This model successfully imitates the decision making process of the transport support officers. As a result of the research, it is shown that the suggested adaptable fuzzy system, which has ability to learn, has a possibility to imitate the decision making process of transport support officers and to show the level of competence comparable with the level of their competence.