Peter Knirsch

Simple Modules for GRACE

The language Grace is being proposed for specifying and programming in a graph-centered environment based on graph transformation. Emphasis in the design of Grace is laid on modular structuring of programs. In this paper we present a simple kind of modules for the language which encapsulate rules and graph transformation units, and export some of them. The local transformation units define compound graph transformations that call rules and other transformation units which are either defined by the surrounding module, or imported from other modules. The interleaving semantics of modules specifies a binary relation on graphs for each exported item.

Distributed Graph Transformation Units

Transformation units are a structuring principle for graph transformation systems. In this paper we introduce distributed transformation units that can be used to model distributed graph transformation systems. A distributed transformation unit consists of a set of local transformation units which are connected via interface units. Semantically, a distributed transformation unit transforms distributed graphs consisting of a set of local graphs connected via interface graphs, in such a way that every local graph with its interfaces is transformed by a local transformation unit. Since the interface graphs can be modified concurrently by various local transformation units we introduce a concurrent semantics of transformation units. The presented concepts are illustrated with a running example of a simple game where two different actors access a common game board randomly.

Autonomous Units: Basic Concepts and Semantic Foundation

Today, most data processing systems and most logistic systems comprise various, possibly distributed, components. These components typically act autonomously, but they may also communicate and interact with each other, spontaneously linking up to form a network. These components do not necessarily need to be stationary. Sometimes they even move or are carried around. Although the components act autonomously, the task to be solved is handled by their interaction and the system as a whole. In this paper the concept of autonomous units for modeling such systems is proposed. Autonomous units form a community with a common environment, in which they act and which they transform. Autonomous units are based on rules, the applications of which yield changes in the environment. They are also equipped with an individual goal, which they try to accomplish by applying their rules. A control condition enables autonomous units to select at any time and in any situation the rule that should actually be applied from the set of all applicable rules.

Undecidable Control Conditions in Graph Transformation Units

Graph transformation units are an approach-independent concept for programming by applying rules and imported transformation units to graphs, starting in an initial and ending in a terminal graph. This transformation process has to obey a so-called control condition, i.e. the device to select how rules or imported transformation units are to be combined in the transformation process executed by the unit. While the other parts of a unit may simply be required to be computable, this is too restrictive for control conditions. In this paper, we show that the semantics of certain control conditions is in general undecidable already when a single imported transformation unit occurs in the condition, and discuss the consequences for programming with graph transformation units.

Autonomy in Software Systems

Looking at the whole logistic network, the structure of logistic processes becomes increasingly complex. Especially in transport logistics, atomisation of transportation processes, multimodal transport chains, international competition, changing ecological and legal constraints along with congestion of traffic infrastructure lead to highly dynamic and complex logistic processes that are difficult to plan (in advance).

Autonomous Units for Communication-based Dynamic Scheduling

Logistics has to deal with dynamics and uncertainties. In order to cope with such problems we introduce a communication-based approach built on distributed autonomous systems. In this work graph transformation with autonomous units is used as a rule-based instantiation of multi-agent systems to model logistic systems. The approachwill be presented using a scenario taken from transport logistics. Here loads which have to be transported are queued in order of their arrival but scheduled for further transportation according to their own constraints. In the paper we propose a negotiation between loads and the respective truck based on payment of transportation rates.

Graph Transformation Modules and Their Composition

In this paper, we investigate the notion of transformation modules as a structuring principle for the specification of graph transformation systems which provide a collection of operations on graphs. Based on the notion of transformation units, a concept that allows to specify binary relations on graphs, a transformation module consists of a set of transformation units. To be able to distinguish between hidden and public operations, a module has an export interface. Moreover, there may be an import interface and a formal parameter. The import interface allows the use of transformation units which are known in the environment of a module. The formal parameter consists of formal parameter units which specify operations on graphs in a loose way. A formal parameter unit may be instantiated by an exported transformation unit of another module through module composition.

Flexible Mass Customisation: Managing Its Information Logistics Using Adaptive Cooperative Multi-agent Systems

Since the last decade, the consideration of customer demands is of increasing importance for manufacturing enterprises. This situation provides a great opportunity to small and medium-sized enterprises (SME) to improve their competitiveness within the global economy. Due to their proximity to the customer, flat management structures and the resulting flexibility, SMEs are able to implement an efficient consumer response (ECR). However, modern industrial products are often characterised by a high complexity of design, functionality, necessary manufacturing and assembly processes. The complexity of modern products may overextend the available skills, knowledge and capacities of a single SME, or it may force this SME to spend great efforts in research and development activities to meet the customer’s requirements.

Semantics of Visual Models in a Rule-based Setting

In this paper, some fundamental aspects of the semantics of rule-based systems are sketched and related to the semantics of visual models. A rule-based system comprises a set of rules and some control conditions including descriptions of initial and terminal configurations. Semantically, the rules specify a binary relation on configurations of some kind by means of rule applications which are restricted according to the control conditions. As visual models are usually represented by diagrams, graphs or similar configurations, the rule-based setting can be employed to provide visual models with semantics.

Mass Customization als Chance für KMU: Kooperative Agenten für die Informationslogistik

Die Technologie der kooperativen Agentensysteme eroffnet kleinen und mittleren Unternehmen (KMU) die Moglichkeit der kundenindividuellen Massenproduktion (Mass Customization) durch effizientes Management der Informationslogistik. Die einzelnen Agenten reprasentieren Unternehmen oder auch Funktionsbereiche in diesen. Die zwischen Unternehmen auszutauschenden Informationen konnen zum grosten Teil durch Agenten automatisch kommuniziert und verhandelt werden. Verteilte Planungs-, Fertigungs- und Montageprozesse werden so in Kooperationen koordiniert. In diesem Beitrag wird ein Konzept am Beispiel „IntaPS“ diskutiert, dass KMU die Chance eroffnet, mit Hilfe einer optimierten Informationslogistik das Potential der kundenindividuellen Massenproduktion auszuschopfen.