Artificial Intelligence

In the planning phase of cyclically interlinked production lines neural networks are used to learn and forecast the characteristics of these systems. To increase the results of learning and to accelerate the training this paper presents a method, based on genetic algorithms, that reduces the attributes to describe the behaviour of the production lines.

The realization of cognitive technical systems deals with the implementation of a knowledge representational level inside technical systems. Hence, cognitive functions and processes (planning, learning etc.) can be used. Especially, the realization of learning (in contrast to adaptation) is the key for novel applications. In application, fields dealing with the guidance of complex systems which can not partially or completely be performed by human operators depending on the system itself, promise a new quality of automation.

In an increasingly turbulent environment, convincing methods of production planning and control are needed. Many of the necessary decisions are made at shop-floor-level. They depend on the knowledge and the abilities of the workers to react on unpredictable impact and hence are not explicitly described. For a realistic, concomitant plant simulation, however, it is important, to model the control strategies as exactly as possible. This paper presents a method to identify applied control strategies by adopting artificial neural networks to data from the operating and machine data logging.

Development projects are prone to continuous changes, whether because of additional insights or changes of requirmenents. To execute changes in a controlled and traceable manner, standardized methods and processes are used in product development, often as part of configuration management. The IT-support of configuration management with defined workflows and release mechanisms is provided by product data management systems (PDM). However, these often fail to acknowledge organizational and financial effects of a change on project management. To remedy this deficit, the article proposes an integrated project/macro-product model implemented by a semantic web and an ontology. This integrated model can be used to run metrics to analyse technical and commercial effects of a change thus improving the change management process as a whole.

The development of new complex products is increasingly conducted cooperatively in distributed, heterogeneous organisations in a dynamic environment. To successfully accomplish a project, the coordination of project controlling is of vital importance. Each participant of the development network has individual controlling systems, which cover his information and management needs. For cross-project controlling however, the creation of interoperability between distributed systems is a precondition for the coordination of processes. The article describes a cross-project controlling approach, which fosters integration by semantic modelling.

Turbulent business environments force enterprises to ever faster answers and adaptions in order to secure their competitive ability. Mutability and responseability as crucial success factors are based finally on knowledge, which must be continuously improved and renewed. The article shows starting points and how ontology based knowledge management can be used purposefully for the increase of the mutability in and by enterprises and in particular in the assembly.

Shorter times to market determine the competitiveness of companies in all industries. The necessary approaches to improvement of industrial processes have to focus on the optimisation of the entire product lifecycle. The resulting management and organisational approach is called PLM. The preliminary integration of all data along the lifecycle can only be handled by semantic technologies. In addition, the introduced approach brings sufficient intelligence to capture enterprise knowledge and effective task support for employees.

Flexibility and transformability of modern manufacturing are determined by many different factors. One of them is an adaptable and dynamic planning of manufacturing processes. The planning systems have not only to guarantee a stable, reliable and quick planning, but also to react to changes and disturbances flexibly and adeqautely. One promising concept is the multi-agents technology. In this work mechanisms and methods for such an agent-based planning are presented that can be applied to the turbulent manufacturing environment.

Speed and cost reduction are plainly the factors of success in global competition. But more and more, the increasing amount of information aggravates the efficient access and finding of information from different systems. Semantic technologies provide convincing solutions by integration of process related knowledge models. Field-tested applications from areas like information retrieval, product development and online consulting illustrate further areas of application.

The design and implementation of knowledge-based control systems is always a major challenge, especially if a maximum realistic approach is pursued. Especially, adequately handling large scale, real world problems, often confronts the concerned project members with insolvable tasks. This article outlines the mutual basis of knowledge management and knowledge engineering plus the challenges en route to acquisition. Solutions to formalise needed knowledge are proposed against the backdrop of a currently running research project in the field of supply chain management with multi-agent systems.