The ability to adaptively control manufacturing equipment, both in local and distributed environments, is becoming increasingly more important for many manufacturing companies.
One important reason for this is that manufacturing companies are facing increasing levels of changes, variations and uncertainty, caused by both internal and external factors, which can negatively impact their performance. Frequently changing consumer requirements and market demands usually lead to variations in manufacturing quantities, product design and shorter product life-cycles. Variations in manufacturing capability and functionality, such as equipment breakdowns, missing/worn/broken tools and delays, also contribute to a high level of uncertainty. The result is unpredictable manufacturing system performance, with an increased number of unforeseen events occurring in these systems. Events which are difficult for traditional planning and control systems to satisfactorily manage.
For manufacturing scenarios such as these, the use of real-time manufacturing information and intelligence is necessary to enable manufacturing activities to be performed according to actual manufacturing conditions and requirements, and not according to a pre-determined process plan. Therefore, there is a need for an event-driven control approach to facilitate adaptive decision-making and dynamic control capabilities.
Another reason driving the move for adaptive control of manufacturing equipment is the trend of increasing globalization, which forces manufacturing industry to focus on more cost-effective manufacturing systems and collaboration within global supply chains and manufacturing networks. Cloud Manufacturing is evolving as a new manufacturing paradigm to match this trend, enabling the mutually advantageous sharing of resources, knowledge and information between distributed companies and manufacturing units. One of the crucial objectives for Cloud Manufacturing is the coordinated planning, control and execution of discrete manufacturing operations in collaborative and networked environments. Therefore, there is also a need that such an event-driven control approach supports the control of distributed manufacturing equipment.
The aim of this research study is to define and verify a novel and comprehensive method for adaptive control of manufacturing equipment in cloud environments.
The presented research follows the Design Science Research methodology. From a review of research literature, problems regarding adaptive manufacturing equipment control have been identified. A control approach, building on a structure of event-driven Manufacturing Feature Function Blocks, supported by an Information Framework, has been formulated. The Function Block structure is constructed to generate real-time control instructions, triggered by events from the manufacturing environment. The Information Framework uses the concept of Ontologies and The Semantic Web to enable description and matching of manufacturing resource capabilities and manufacturing task requests in distributed environments, e.g. within Cloud Manufacturing. The suggested control approach has been designed and instantiated, implemented as prototype systems for both local and distributed manufacturing scenarios, in both real and virtual applications. In these systems, event-driven Assembly Feature Function Blocks for adaptive control of robotic assembly tasks have been used to demonstrate the applicability of the control approach. The utility and performance of these prototype systems have been tested, verified and evaluated for different assembly scenarios.
The proposed control approach has many promising characteristics for use within both local and distributed environments, such as cloud environments. The biggest advantage compared to traditional control is that the required control is created at run-time according to actual manufacturing conditions.
The biggest obstacle for being applicable to its full extent is manufacturing equipment controlled by proprietary control systems, with native control languages. To take the full advantage of the IEC Function Block control approach, controllers which can interface, interpret and execute these Function Blocks directly, are necessary.
United Kingdom: De Montfort University , 2018. , p. 232
Function Block, Adaptive control, Robot-Control-as-a-Service, Cloud Manufacturing