The transport demand is increasing year by year. This is because of the increase in production and consumption. The longer the supply chain is, the more likely it is to be disrupted, as all operations involve risks. However, risk in the context of the supply chain and its management has recently been discussed. Unfortunately, no industry is immune to predictable and unpredictable disruptions that affect losses (e.g. loss of goods). From the point of view of carriers, it would be important to be able to predict the occurrence of, for example, damage to the goods. The article focuses on the use of machine learning methods to predict the risk of damage to goods (such as electronics, household appliances or telephones/computers) in road transport. The research used five intelligent methods such as: logistic regression; support vector machine (SVM); decision tree; naive Bayesian classifier; AdaBoost. The aim of the paper is to present the concept and the above-mentioned methods of machine learning, measures assessing the performance of models and the results related to the conducted research. The set goal determined the choice of the research methods – literature analysis and programming were used. The last part of the article presents the results obtained from the analysis of five models. The research established that AdaBoost has the best predictive ability.

In recent years, more and more attention has been paid to the use of Big data technology, machine learning and AI. Enterprises strive for a competitive advantage through the appropriate use of data analytics. Big data can be used in many different industries, e.g. in the transport or medical industry, and potentially in all of them. A huge problem in the supply chain is the risk of delay, which may be influenced by many factors, including illegible label on the package, lack of warehouse workers or congestion in cities. The article focuses on the use of Big Data technology to detect the risk of delays in the supply chains of medicinal products. Its purpose is to present the concept of Big Data, Big Data architecture for the drug supply and to present the results of research related to the prediction of the risk of delays in its implementation in a real enterprise. The set goal determined the choice of the following research methods: analysis of literature and the use of modeling, which allowed to design and implement the architecture for the drug supply chain to collect data in the studied enterprise. The last part of the article presents a logistic regression model for predicting delays in the supply chain of medicinal products. The research established that the model has a high predictive ability.