Bankruptcy represents one of the important today's business problem, the occurrence of which can endanger not only one company, but also its numerous business partners (suppliers, banks,...). This paper tries to analyze bankruptcy forecasting models and identify the key causes of the occurrence of this phenomenon. Therefore, the research problem of this paper is the problem of determining and analyzing financial indicators that explain the probability of the occurrence of business failure of a company, i.e. bankruptcy. The goal of this thesis is to find out whether there is a possibility of predicting bankruptcy, respectively the goal is to find and analyze the causes that indicate that there is a potential danger of bankruptcy of a certain company or a certain industry. Attention is paid to the hypothesis that reads: Indicators of profitability, liquidity, indebtedness and activity explain the probability of business failure, in the form of bankruptcy. The sample included 80 companies from the processing industry in Croatia, of which 40 are bankrupt and 40 are healthy companies. Small, medium and large enterprises were observed, and micro enterprises were not taken into consideration. The accuracy of the entire model, for the observed period (t-1), is 94.1%, respectively at the level of successful companies in the processing industry it is 95.0%, more than for companies in bankruptcy (92.90%). Looking at the other two periods within this research, the accuracy has decreased, but is still at a satisfactory level. For the period (t-2), the overall prediction accuracy is 68.90%, and for the period (t-3) it is 75.70%, which represents this overall model as successful. Accordingly, the hypothesis set at the beginning of this paper is accepted based on the accuracy of the statistical analyses.