Analysis of the factors affecting the export performance in TR 81 Level 2 region with decision tree modeling

With the increase in integrated regional applications projects in Smart Logistics activities, advantages and opportunities have increased for companies operating in export and import activities in the region. Investments made in the TR81 Level 2 region are expected to be a gateway to new markets for companies that have not engaged in export activities before. In the study, it was aimed to analyze the criteria affecting export performance with a 5-point Likert-structure survey conducted on 74 companies (63 exporters, 11 non-exporters) operating in the provinces of Bartın, Karabük and Zonguldak, which are called TR81 Level 2. When the findings of the section consisting of 8 main criteria and 56 sub-criteria are examined, it is understood that they have high participation in the criteria of R&D activities, customer satisfaction criteria and personnel status. According to the results of the Mann-Whitney U test applied between exporting and non-exporting firms, no significant differences were found in the general approaches to factors. However, it was revealed as a result of the analyzes that there were significant differences in the items. It has been observed that the approaches of exporting and non-exporting firms to 8 main factors are similar. It has been determined that exporting companies have a high level of participation in the export strategy, organization and desire for success. Using the J48 algorithm in data mining decision tree modeling, 7 scenarios were created. With the scenarios, it is aimed to create road maps that are planned to help companies and managers who want to export. Variables such as the manager's overseas experience, the firm's turnover status, quality targets, the firm's equity status, R&D factor, customer satisfaction factor, advertising and promotional activities are frequently included in the models. It was concluded that the variables are the determinants of exporting.