A importância da ciência de dados como facilitador de tomadas de decisão na gestão das cadeias de abastecimento

Abstract

This research studies the phenomenon of Digital Transformation (DT) by investigating a SYSTEM consisting of four disciplines: Supply Chain Management (SCM), Knowledge Management, Artificial Intelligence (AI) and Data Science (DSc). The hypotheses of the research are: (H1) the profiles of absorption of the knowledge of the SYSTEM, in different countries, are similar when there is a common knowledge base absorbed in these countries; and (H2): the profiles of application of knowledge about the SYSTEM, among countries that share a common basis of this knowledge, are similar. The study categorized into domains the factors that influence DT. A method was developed that combines questionnaire formulation techniques and sampling of very large populations, with more than 5,000 individuals. A questionnaire was sent to 13,479 rigorously selected individuals to study a target population of 37,592 professionals registered on LinkedIn. The 844 valid responses from 60 countries show that companies worldwide make decisions in SCM based on DSc processes that use AI practices to address the Big Data generated by Stakeholders. We conclude that the stage of absorption and application of the measured knowledge is similar in the countries studied. In the context of the Knowledge Economy society, without rigid barriers capable of preventing the transfer of knowledge, the implications of this mechanism include the consolidation of a common knowledge base between countries and the consolidation of similar levels of maturity related to the application of this knowledge. An additional contribution is that the method, which follows a clear sequence of tasks, provides the expected values for the Sampling Rate, Return Rate and Sample Error in surveys with very large target population on LinkedIn. Consequently, it is possible to reverse situations such as the drop-in response rate, low selectivity of the target population and little representativeness of very large populations. The method can be applied in surveys, transdisciplinary or not, of various areas of science, strengthening the credibility of the conclusions elaborated from the results obtained. Additionally, the Sampling and Results Matrix (SRM), a key element of the methodology, is a valuable tool for the academic benchmark, organizing the same indicators present in different studies and stimulating the standardization of interpretations of results by the scientific community.