Applying Artificial Intelligence in MAPNA Multimodal Transportation Company in Order to Find the Optimal Rail Load

The ever-increasing demand in the field of rail freight transportation and the competitiveness of the price of goods transportation and rail transportation fares compared to other transportation sectors in the country is one of the most important reasons for the development and investment as much as possible in this sector of transportation. Different criteria in the field of rail cargo transportation, both by the owners of the goods and by the rail transportation companies licensed to issue a rail bill of lading, have been discussed and investigated in recent years. The main purpose of this article is to apply the use of artificial intelligence in MAPNA multimodal transportation company as one of the important companies in the field of rail cargo transportation, in order to find the optimal load. In this article, the implementation of multi-layer artificial neural networks and the description of the implementation codes were done through the Python programming language and after evaluating and adjusting the model parameters, the weights of the network were stored by the pickle library that can be used to estimate new data. After checking 89275 bills of lading available in the railway network, the information of the mentioned bills of lading was entered into the simulated program, the main formula of which was formulated by industry experts and railway experts. The results of the optimal rail load from among the issued bills of lading show more income and lower costs.