Workflow Systems are a means to automate business processes, including modeling, simulation, monitoring, and execution. The idea is to provide a high-level tool-chain to allow non-programmers to describe and drive their business cases (processes) in a graphical way.
From a technical point of view, a process is a series of steps (tasks) and decisions. Each task represents work that has to be done, by either a machine in a factory, a human, or a computer system. The granularity of a process can be very diverse: from the controlling of individual machines in a manufacturing line, to just delegating tasks to humans or web services to control the interaction between them.
Whether an individual machine is controlled directly through a process, or the process just triggers a control program (black box), is related to the level of flexibility that is needed. For example if customization of products is an important business goal, exposing fine grained control over the physical manufacturing process to non-factory floor staff and non-programmers can make all difference.
Another important issue is adaption and evolution of processes: through the explicit description of (production) processes, the knowledge about what is going on the factory floor is made transparent. When changes to the production process are necessary, these changes can be accurately planned and executed, as the consequences to the factory floor are known beforehand. This again is crucial e.g. for flexible customization of products.
In order to support a variety of use cases with state-of-the-art cloud-based process technology, the Workflow Systems and Technology Group at the university of Vienna develops tools and methodologies to support dynamic process execution, simulation, and monitoring.
Our own open source Cloud Process Execution Engine CPEE, can run on embedded systems, as well as in cloud deployments, and has proven its maturity in various settings, ranging from the health-care sector to the utilization in virtual factories.
The WST group was a partner in the ADVENTURE FP7 EU project (GA no 231396.). This projects provides the central piece of a framework that provides the tools to combine factories in a pluggable way to manufacture a particular product. This includes the creation of manufacturing processes, finding partners and real-time monitoring of the processes. Examples for the the integration in real-world manufacturing processes include:
- Collecting sensor data and integrating it with processes to allow for automatic decisions, e.g. when to shut down a production line for maintenance, when to reorder products, or in case of disasters.
- Guaranteeing a secure exchange of information between partners in a supply chain, as well providing a coherent security model for data exchange between multiple parties (i.e. the process describes the what, when and how data is exchanged).
- Real-time monitoring of what is going on in multiple production lines, as well as aggregating and exposing crucial information such as stock levels and OEE (overall equipment efficiency) to factory managers.
The Workflow Systems and Technology Group also specializes in forecasting and simulation of processes. Several approaches have been explored:
- For predicting the performance of processes (in the immediate future) that strictly deal with factory internal resources, it is possible to rely on the analysis of historical data, collected from events disseminated during process execution.
- For predicting the feasibility of processes that deal with external partners it is better to collect and aggregate forecasting (business intelligence) data from partners.
The Workflow Systems and Technology Group aims at showcasing how to allow for fast decisions and dynamic adaptations of production processes, through tight integration of the factory floor with powerful process technology.
|Forschungsgruppe Workflow Systems and Technology|