Monitoring data quality in corporate r&d management processes using fuzzy systems

In today's rapidly changing economic and technological environment, sustainable growth and competitive advantage depend on the right selection and management of R&D projects. R&D activities are of strategic importance to companies in order to increase the efficiency of production processes, improve existing products and develop new products and services. However, project selection is a complex decision-making process that requires the review of multiple data quality standards. In this context, the knowledge and experience of expert decision makers play a crucial role in the evaluation of projects. However, due to the multiple criteria and uncertainties involved in this process, the use of advanced decision support systems such as fuzzy logic is required. This thesis details the R&D project selection process carried out in an R&D unit approved by the Ministry of Industry and Technology in an industrial electronics company operating in Turkey. The focus of the research is to propose and implement a model using a fuzzy logic method based on project types and data quality criteria. In this context, opinions from five different experts have been obtained for projects to be evaluated based on five different data quality criteria (Data Accuracy, Data Completeness, Data Consistency, Data Timeliness, Data Precision). The projects evaluated by the experts have been examined under four different project types (P0, P1, P2, P3), each with different characteristics. Expert opinions collected through forms created for each project type were applied as input to fuzzy logic models. Acceptable data quality criteria levels for each project type were passed through fuzzy logic models and minimum acceptance levels were determined. For the purpose of project selection, three different projects from each project type were examined by experts and data quality levels were determined. A total of twelve projects were passed through the fuzzy logic model of their project type, the results were obtained and compared with the minimum acceptance level. In the evaluation, three of twelve different projects were found to be suitable in terms of data quality and it was decided to include them in the project plan. The results have demonstrated the applicability and effectiveness of the proposed model, facilitating decisions on the inclusion of projects selected on the basis of certain criteria in the R&D planning process. The proposed method represents an important resource for the selection and management of internal innovation and R&D projects.