Formation of Complex Company Evaluation Method through Neural Networks Based on the Example of Construction Companies ́ Collection

The main goal of this contribution is to create a model through the use of neural networks, which will be able to predict the company ́s ability to survive a prospective financial crisis. Design/methodology/approach: The contribution examines the basic data on companies coming from the Albertina database. The collection includes both financial and non-financial indicators of all construction companies in the Czech Republic within the period of 2008 to 2014. Theobject is to find an artificial neural network which can classify each company based on the input data into one of the following four groups: the company is not on the verge of bankruptcy, collapse in the given year, collapse in the following two years and collapse in future. Findings:Three neural networks are given and described, proving positive results. With regard to the usefulness ofthemodel and minimal deviations from the spare two models which were going to be taken into account, the MLP 15:15-54-66-4:1 model proves the best results. Through this network theCzech construction companies ́ ability to survive a possible distress is consequently evaluated. Research/practical implications:Theevaluation of financial and non-financial indicators is a necessary part of modern companies. Artificial neural networks are able to conduct non-linear statistical modelling and offer acompletely suitable alternative to individual financial indicators within complex methods ofthis evaluation. Originality/value:Using artificial intelligence, the contribution creates a bright new methodology for comprehensive business evaluation, which includes both financial and non-financial indicators. The result could be used for practical comprehensive evaluation of construction companies in the Czech Republic