In the last decades, the increasing complexity of industrial information technology has led to the emergence of new trends in manufacturing. Factories are using multiple Internet of Things (IoT) platforms to harvest sensor information to improve production. Such a transformation contributes to efficiency growth and reduced production costs. To deal with the heterogeneity of the services within an IoT system, Service-Oriented Architecture (SOA) is referred to in the literature as being advantageous for the design and development of software to support IoT-based production processes. The aim of SOA-based design is to provide the leverage to use and reuse loosely coupled IoT services at the middleware layer to minimise system integration problems. We propose a system architecture that follows the SOA architectural pattern and enables developers and business process designers to dynamically add, query or use instances of existing modular software in the IoT context. Furthermore, an analysis of utilization of modular software that presents some challenges and limitations of this approach is also in the scope of this work.

The notion of Cyber-Physical Systems (CPS) is proposed by the National ScientificFoundation to describe a type of systems which combine hardware and software components andbeing the next step in development of embedded systems. CPS includes a wide range of researchtopics ranging from signal processing to data analysis. This paper contains a brief review of the basicinfrastructure for CPS including smart objects and network aspects in relation to TCP/IP stack. AsCPS reflect the processes of the physical environment onto the cyber space, virtualisation as animportant tool for abstraction plays crucial role in CPS. In this context paper presents the challengesassociated with mobility and vritualisation; accordingly three main types of virtualisation, namelynetwork, devices and applications virtualisation are presented in the paper. These aspects are tightlycoupled with security and safety issues. Therefore, different threats, attack types with correspondingsubtypes and possible consequences are discussed as well as analysis of various approaches to copewith existing threats is introduced. In addition threat modelling approaches were also in scope of thiswork. Furthermore, needs and requirements for safety-critical CPS are reviewed. Thus the mainefforts of this paper are directed on introducing various aspects of the CPS with regard to securityand safety issues.

The concept of Cyber-Physical Systems (CPS) is the next stage of Embedded Systems development and involves tied integration of physical, cyber, and communication components. The rapid growth in number of interconnected devices and objects requires new governance, aggregation and structuring mechanisms for objects themselves and data generated by them. Furthermore, new organizational structures are needed to cope with scalability and integration of new devices/objects without disturbing already functioning devices/objects. As objects become smarter, with increasing computational capabilities, the implementation of collaboration mechanisms among them becomes relevant. This paper discusses current trends and challenges for implementation of collaborative CPS. Moreover, a smart home scenario is presented, in order to illustrate such challenges.

The next evolutionary technological step in the industry presumes the automation of the elements found within a factory, which can be accomplished through the extensive introduction of automatons, computers and Internet of Things (IoT) components. All this seeks to streamline, improve, and increase production at the lowest possible cost and avoid any failure in the creation of the product, following a strategy called “Zero Defect Manufacturing”. Machine Learning Operations (MLOps) provide a ML-based solution to this challenge, promoting the automation of all product-relevant steps, from development to deployment. When integrating different machine learning models within manufacturing operations, it is necessary to understand what functionality is needed and what is expected. This article presents a maturity model that can help companies identify and map their current level of implementation of machine learning models.