Breakeven point analysis by using artificial neural networks for management accounting: An application in dangereous goods transportation

In this study, the break-even point for an enterprise subject to the ADR (European Agreement Dangerous Goods Goods Road), an international regulation, was calculated using the artificial neural network method. Due to the calculation of the break-even point for the future of the related production operation subject to the ADR Agreement, data production was done according to the uniform distribution function within a certain interval based on actual production data of one month. The artificial neural network method, which is used as an alternative analytical technique to the ADR Agreement, an international regulation on dangereous goods and classical cost-volume-profit analyzes, constitutes the essence of the thesis work. Artificial neural network was constructed by using multi layer forward feed back propagation algorithm in thesis study. This network structure consists of an input layer, three hidden layers with 10 neurons in each hidden layer, and an output layer. 85% of the data set was used for network training and 15% was used for simulating network training. As a result of the artificial neural network simulation performed for the training, the breakeven point sales forecast performance was calculated as 86.66% and the breakeven point forecast performance was 70%. The results of the analysis show that the multi-layer artificial neural network is successful method.