No abstract

No abstract

Amid the current industrial revolution, a total disruption of the existing production lines may seem to be the easiest approach, as the potential possibilities seem limitless when starting from the ground up. On the business side, an adaptation of existing production lines is always a preferred option. In support of adaptation as opposed to disruption, this paper presents a new approach of using production process orchestration in a smart factory, discussed in an industrial case-study example. A proposed smart factory has the Orchestrator component in its core, responsible for complete semantical orchestration of production processes on one hand, and various factory resources on the other hand, in order to produce the desired product. The Orchestrator is a complex, modular, highly scalable, and pluggable software product responsible for automatised planning, scheduling, and execution of the complete production process. According to their offered capabilities, non-smart and smart resources—machines, robots, humans—are simultaneously and dynamically assigned to execute their dedicated production steps.

One of the main challenges of Industry 4.0 is the adaptation of existing production lines. Robots are substituting human workers in modern smart factories, as they are much more suitable for repetitive tasks. In contrast to that, Industry 4.0 predicts a high rise in product customization. The total disruption of the current factories, although the easiest solution, is not welcomed by the traditional industry stakeholders. To offer adaptation rather than disruption, and to promote man-machine collaboration rather than complete substitution of the human workforce, we present a Digital Factory solution capable of orchestrating different types of resources -humans, machines, robots -according to their capabilities. The core component of the solution is a real-time Orchestrator that orchestrates factory resources in order to produce the desired product. Orchestrator is a complex, modular, highly scalable, and pluggable software, responsible for dynamical matching, scheduling, and executing of production steps, allowing high customization and lot-size-one production.

Technological advances and increasing customer need for highly customized products have triggered a fourth industrial revolution. A digital revolution in the manufacturing industry is enforced by introducing smart devices and knowledge bases to form intelligent manufacturing information systems. One of the goals of the digital revolution is to allow flexibility of smart factories by automating shop floor changes based on the changes in input production processes and ordered products. In order to make this possible, a formal language to describe production processes is needed, together with a code generator for its models and an engine to execute the code on smart devices. Existing process modeling languages are not usually tailored to model production processes, especially if models are needed for automatic code generation. In this paper we propose a research on Industry 4.0 manufacturing using a Domain-Specific Modeling Language (DSML) within a Model-Driven Software Development (MDSD) approach to model production processes. The models would be used to generate instructions to smart devices and human workers, and gather a feedback from them during the process execution. A pilot comparative analysis of three modeling languages that are commonly used for process modeling is given with the goal of identifying supported modeling concepts, good practices and usage patterns.