Integrating renewables into the grid introduces price fluctuations that stem from their inherent volatility. To mitigate this, adopting an energy-aware production planning strategy benefits factories and the grid. By aligning production processes with periods of abundant renewable energy and lower costs, factories can actively reduce their energy expenditures while simultaneously smoothing grid strain. However, such a transition requires expensive retrofitting of production planning systems. This paper proposes an approach to transparently integrating energy awareness into an existing Manufacturing Execution System (MES), resulting in a cost reduction without the need to retrofit the MES.

HVAC systems are characterized by rigid and complex internal behaviour that limits the options for efficiency optimization. KI4HVACs project aims to apply machine learning to improve office building HVAC operation and maintenance efficiency. Reinforcement learning is used for energy use optimization and supervised learning for preventive maintenance. The models are wrapped into an MPC based and a decision tree algorithm for preventive maintenance. The system uses the TRNSYS18 simulation environment that matches an actual office building with multiple zones.

The growing emphasis on environmental sustainability, driven by regulations such as Europe's Carbon Border Adjustment Mechanism (CBAM) and increased consumer awareness, is a notable trend. It is estimated that manufacturing energy consumption accounts for 24 % of global greenhouse gas emissions. As demand for renewable energy rises, current market mechanisms are unable to guarantee a carbon-free energy supply. This paper presents an approach that integrates energy availability and product carbon footprint (PCF) into production planning. The necessity for factories and energy suppliers to adapt their operations is highlighted. To ensure the optimal utilization of available energy resources, it is essential that factories implement production schedules that are responsive to fluctuations in supply. Energy suppliers must provide accurate long-term production profiles and employ anomaly detection for short-term adjustments. The proposed solution is designed to align energy consumption with renewable energy availability and PCF.

Production planning is not normally concerned with energy costs, but with the increasing share of volatile renewable energy, energy prices are becoming volatile, too. Production planning will ultimately have to take this into account. The problem here is in the uncertainly of price predictions and the associated risk. The present work investigates how to integrate dynamic energy price prediction uncertainty into the optimization of energy consumption in an industrial production setting. To that end, we incorporate the cost of prediction errors into the predicted energy prices up to the planning horizon. Such risk-adjusted prices are then used as the basis for energy-aware production planning. First results show that this concept can handle different error distributions and can be assumed feasible.

Energy consideration in production planning is becoming more important due to concerns over the climate change. However, the uncertainty of energy availability hinders the optimisation of production. The proposed concept uses an iterative approach of adjusting production scheduling in respect to the available energy using a multi-criteria optimisation that not only includes factory utilisation but also the likelihood of the predicted energy availability.

In smart manufacturing, the demand increases to be able to monitor and adjust process execution during production. It has led to a shift towards distributed, modular automation. A solution seems to be to use a cognitive architecture. It turns out that their generality often makes them unsuitable for specific industrial problems. In this paper, we propose a cognition-inspired architecture design for health monitoring tasks. The problem class is represented by a conveyor belt use case. We then discuss to what extent this architecture matches common implementations of cognitive theories by following a generalized cognitive process.

In the context of Industry 4.0, constantly evolving shop floors generate the need for a highly adaptive and autonomous automation system with lean maintenance, minimum downtime, maximum reliability, and resilience. Future Manufacturing Execution Systems (MESs) will be more complex and dynamic as well as distributed physically and logically. This makes it very difficult, if not impossible, for the conventional centralized architectures to effectively control these vibrant Cyber-Physical Production Systems (CPPSs). To address these issues, we propose Self-Aware health Monitoring and Bio-inspired coordination for distributed Automation systems (SAMBA), an architecture which tackles these challenges. SAMBA increases the ability of the system to intelligently adapt to rapidly changing environment and conditions of future CPPSs.

Modern trends in automation require flexible production with an onus on saving energy. Till recently, the focus of production systems has been only saving production costs, but the growing concern over environmental impact of such system has led to a partial change in paradigm. The key for the future generation production system is to successfully integrate the manufacturing environment with energy optimisation. This relies on transparent representation of the building from the point of the Manufacturing Execution System (MES) that is not feasible using conventional centralized approaches to building thermal modeling and optimization. This work focuses on distributing the building automation system in a set of generic cells that encapsulate thermal energy models and optimization algorithms to minimize computational resources required for the system implementation on embedded devices. Combined with a notion of the global overview of the energy optimization with functional aggregation and dynamic clustering concepts, the proposed architecture provides a flexible energy saving solution that can be easily integrated in majority of conventional manufacturing systems as well as related factory buildings without compromising production operations.

Up to now, production systems only concern was to minimize production costs or to optimize the utilization of production resources. But with the increasing energy prices and the growing concern over the environmental impact of production systems (industrial systems consume a quarter of all energy), efficient use of energy in manufacturing environment cannot be ignored any longer. MANUbuilding concept brings together the manufacturing systems requirements with the building automaton concerns over the efficient energy use and develops the local decision making based on dynamically expanding clusters to optimize the energy consumption in a factory.

The function block standard IEC 61499 is gaining popularity as a solution for distributed automation systems. Its event-based concept suggests the use of packet-based networks such as conventional Ethernet as underlying communication infrastructure. For such networks, however, proper Quality of Service (QoS) mechanisms are required to ensure adequate network performance. Ideally, these mechanisms must be configured and controlled from the IEC 61499 system both during engineering and normal operation. This paper addresses the problem of fully integrating Ethernet switches into an IEC 61499 system rather than treating them as transparent network components. An analysis of existing QoS support in available devices reveals a large degree of heterogeneity in the implementations that calls for an abstraction layer. This abstraction is introduced by means of a dedicated QoS function block attached to the device which makes the QoS parameters accessible from the function block application layer thus making the device-specific implementation transparent to the IEC 61499 system developer.

Manufacturing Execution Systems are essential elements for vertical integration in modern automation systems because they provide the link between the centralized enterprise level and the distributed shop floor. Existing centralized MES implementations can however not properly cope with an unpredictable order flow, changes on the shop floor, or the increasing complexity of both. Distribution of the MES functionalities would allow shifting the decision making process from the enterprise down to the field level, resulting in a more flexible manufacturing system. This paper investigates how and to which extent MES functionalities can actually be distributed in different approaches with varying degree of organization and shows what affects system behavior with respect to flexibility and optimization.

Modern trends in manufacturing demand vertical integration between the enterprise and the field control level. Existing Manufacturing Execution System solutions fail to satisfy the requirements of unpredictable order flow and dynamic shop floor; neither can they deal with the increasing complexity of both. Distribution of the MES functionalities would allow shifting the decision making process from the enterprise down to the field level and providing concurrency of execution that results in a flexible system covering all three levels of the automation pyramid. This paper investigates how and to which extent MES function can actually be decentralized and describes a possible implementation based on software agents.

Modern trends in manufacturing demand more flexible and adaptive handling of production order execution. Due to its distributed nature, agent-based systems become a suitable solution to cope with large scale and changeable industrial systems making management and execution of a production order crucial for the system efficiency. PABADISpsilaPROMISE is an agent-based architecture with a dedicated entity for each order that is equipped with mechanisms to fulfill the requirements of the distributed control.

Distributed control systems are emerging as a solution to cope with complexity of production and lack of flexibility of the conventional centralized control system. PABADIS'PROMISE provides an architecture for an agent-based distributed control of the production on all three layers of the automation pyramid. One of the aspects of the approach is a multilevel mechanism for order decomposition that increases flexibility and decreases the complexity of the system.

With the exploding complexity of distributed production, current management concepts cannot feasibly maintain full control over work pieces in the plant. Product identification technologies such as RFID are a solution to retain control. Nevertheless they are only a piece in the puzzle to retrieve the desired information at shop floor level and superordinate levels like ERP systems. This paper analyses the suitability of RFID (radio frequency identification) in agent-based control applications to gain full control over all products and resources at shop floor level. Additionally, an implementation concept for RFID-based product control developed in the EU project PABADIS' PROMISE is presented; the project aims at installing a completely distributed agent-based production control system with full information control.