Big Data-based methods for functional safety case preparation

Aim. The paper aims to overview the opportunities, approaches and techniques of studying ensuring functional safety transportation systems, including those driverless, with use Big Data. It is noted, that modern technology underpins next-generation systems operate in ever-evolving conditions, significant numbers passengers, requires modified control design. With growth agglomerations, many suburban merge urban ones, their traffic intervals are close size metro. Under these there a transition from human-machine automatic characterized by varying degrees automation. Widespread deployment digital telecommunications, process automation remote data collection management under way railway transportation. Variations behaviour as type cyberphysical system cause paradigm shift line-andstaff adaptive fundamentally non-linear variable structure parameters. Methods. Control conventionally assessed for Lyapunov’s stability. In this case, stable can 100% probability be predicted neighbourhood ε-tube. For examined supervised which stability ensured through introduction supervisor algorithm, speaking strict would not correct. idea controlled algorithms extend only ANN, but also other intelligent algorithms. Thus, scope knowledge identified covered relevant regulatory documents methods case preparation. Identifying eliminating abnormal signals such allow defining boundaries set acceptable processes more clearly, thus, some cases, increasing speed decision disabling an entire branch unfavourable scenarios. Results. stricture parameters, examples considered machine learning/Big Data application analysing complex control/management proposes concept artificial neural networks combined model checking. A special attention given elements new subclass networks. Conclusion. Updated requirements defined using intelligence part train schedule autonomous control. That will ultimately developing line research associated operation counterintuitive AI-based estimation learning preparation based on formal verification.