Autoren:  Peter Dadam, Manfred Reichert

This paper gives insights into the ADEPT project. Its target was to develop a next generation process management technology, which is by orders of magnitudes more powerful and flexible than contemporary process management systems. The ADEPT technology should provide advanced features and properties within one system, which seem to exclude each other, but which are required for the support of a broad spectrum of processes: ease-of-use for end users and system developers, high flexibility through the support of non-trivial ad-hoc deviations at the process instance level, quick implementation of process changes through process schema evolution, and correctness guarantees enabling robust execution of implemented processes. This paper describes the background and the real-world cases which motivated our research. It further explains the technological challenges we faced, describes the solutions we elaborated, and discusses the current status of the ADEPT project.

PDF

Autoren:  Peter Dadam, Manfred Reichert, Stefanie Rinderle-Ma, Kevin Göser, Ulrich Kreher, Martin Jurisch

Angeregt durch ein Forschungsprojekt im Bereich klinischer Informationssysteme, wurde von uns Mitte der 90er Jahre das Forschungsprojekt ADEPT gestartet, welches im Bereich Prozess-Management das nahezu Unmögliche anstrebte und mittlerweile auch erreicht hat: Hochgradig flexible Ausführung von Unternehmensprozessen, Realisierung robuster prozessorientierter Anwendungen "per Konstruktion" sowie ein für alle Anwendergruppen (Prozess-Implementierer, Systemadministratoren, Endbenutzer) einfach zu benutzendes System. Dieser Beitrag beschreibt die Hintergründe des ADEPT-Projekts sowie unsere Motivation für die gesteckten Ziele, die von uns verfolgte Vision und deren vollständige Umsetzung in der nunmehr verfügbaren AristaFlow® BPM Suite.

PDF

Autoren:  Alena Hallerbach, Thomas Bauer, Manfred Reichert

When engineering process-aware information systems (PAISs) one of the fundamental challenges is to cope with the variability of business processes. While some progress has been achieved regarding the configuration of process variants, there exists only little work on how to accomplish this in a correct manner. Configuring process variants constitutes a non-trivial challenge when considering the large number of process variants that exist in practice as well as the many syntactical and semantical constraints a configured process variant has to obey in a given context. In previous work we introduced the Provop approach for configuring and managing process variants. This paper picks up the Provop framework and shows how it ensures correctness of configurable process variants by construction. We discuss advanced concepts for the context- and constraintbased configuration of process variants, and show how they can be utilized to ensure correctness of the configured process variants. In this paper we also consider correctness issues in conjunction with dynamic variant re-configurations. Enhancing PAISs with the capability to correctly configure process models fitting to the given application context, and to correctly manage the resulting process variants afterwards, will enable a new quality in PAIS engineering.

PDF

Autor: Martin Bader

During the last years, the genomes of more and more species have been sequenced, providing data for phylogenetic reconstruction based on genome rearrangement measures. A main task in all phylogenetic reconstruction algorithms is to solve the median of three problem. Although this problem is NP-hard even for the simplest distance measures, there are exact algorithms for the breakpoint median and the reversal median that are fast enough for practical use. In this paper, we extend this approach to the transposition median as well as to the weighted reversal and transposition median. Although there is no exact polynomial algorithm known even for the pairwise distances, we will show that it is in most cases possible to solve these problems exactly within reasonable time by using a branch and bound algorithm.

PDF

Autoren:  Barbara Weber, Andreas Lanz, Manfred Reichert

Formal specification and operational support of time constraints constitute fundamental challenges for any enterprise information system. Although temporal constraints play an important role in the context of long-running business processes, time support is very limited in existing process management systems. By contrast, different kinds of planning tools (e.g., calendar systems and project management tools) provide more sophisticated facilities for handling task-related time constraints, but lack an operational support for business processes. This paper presents a set of 10 time patterns to foster the systematic comparison of these different technologies in respect to time management. The proposed patterns are all based on empirical evidence from several large case studies. In addition, we provide an in-depth evaluation of selected process management systems, calendar systems and project management tools based on the suggested patterns. The presented work will not only facilitate comparison of these different technologies in respect to their support of time constraints, but also make evident that their integration offers promising perspectives in respect to time support for long-running business processes.

PDF

Autoren:  Stefanie Rinderle-Ma, Manfred Reichert

Enabling changes at both process type and process instance level is an essential requirement for any adaptive process-aware information system (PAIS). Particularly, it should be possible to migrate a (long-)running process instance to a new type schema version, even if this instance has been individually modified before. Further instance migration must not violate soundness; i.e., structural and behavorial consistency need to be preserved. Compliance has been introduced as basic notion to ensure that instances, whose state has progressed too far, are prohibited from being migrated. However, this also excludes them from further process optimizations, which is not tolerable in many practical settings. This paper introduces a number of strategies for coping with non-compliant instances in the context of process change such that they can benefit from future process type changes on the one hand, but do not run into soundness problems on the other hand. We show, for example, how to automatically adjust process type changes at instance level to enable the migration of a higher number of instances. The different strategies are compared and discussed along existing approaches. Altogether, adequate treatment of non-compliant process instances contributes to full process lifecycle support in adaptive PAIS.

PDF

Autoren:  H. A. Kestler, B. Lausen, H. Binder, H.-P. Klenk, F. Leisch, M. Schmid (eds)

41. Arbeitstagung der Arbeitsgruppen Statistical Computing (GMDS/IBS-DR), Klassifikation und Datenanalyse in den Biowissenschaften (GfKl) und dem Arbeitskreis Computationale Statistik (ÖSG). 28.06.-01.07.2009, Schloss Reisensburg (Günzburg).

PDF

Autoren: Ulrich Kreher, Manfred Reichert, Stefanie Rinderle-Ma, Peter Dadam

Ein Prozess-Management-System (PMS) muss umfangreiche Funktionen für die Steuerung, Verwaltung und Änderung von Geschäftsprozessen bieten. Um breit einsetzbar zu sein, ist eine performante Ausführung dieser Funktionen unverzichtbar, insbesondere bei groÿer Anzahl von Prozessinstanzen. In diesem Beitrag untersuchen wir einen wichtigen Performanzaspekt von PMS, die eziente Speicherrepräsentation von Vorlagen- und Instanzdaten. Dazu setzen wir ein logisches Prozessmetamodell in programmiersprachenunabhängige technische Datenstrukturen um und zwar sowohl für Primär- als auch Sekundärspeicher. Wichtigstes Ziel ist die Minimierung des Primärspeicherbedarfs. Nur dadurch lässt sich zeitaufwendiges Ein- und Auslagern der zur Laufzeit benötigten Vorlagen- und Instanzdaten minimieren. Wir diskutieren grundlegende Realisierungsalternativen für die Speicherrepräsentation dieser Daten, und gehen darüber hinaus auf fortschrittliche Speicherkonzepte, wie die Clusterung von Prozessinstanzen, ein. Die vorgestellten Konzepte sind aktuell im ADEPT2-PMS umgesetzt.

PDF

Autoren: Holger Dammertz, Alexander Keller, Hendrik P.A. Lensch

We present a physically-based progressive global illumination system that is capable of simulating complex lighting situations robustly by efficiently using both light and eye paths. Specifically, we combine three distinct algorithms: point-light-based illumination which produces low-noise approximations for diffuse inter-reflections, specular gathering for glossy and singular effects, and a caustic histogram method for the remaining light paths. The combined system efficiently renders low-noise production quality images with indirect illumination from arbitrary light sources including inter-reflections from caustics and allows for simulating depth of field and dispersion effects. Our system computes progressive approximations by continuously refining the solution using a constant memory footprint without the need of pre-computations or optimizing parameters beforehand.

PDF

Autoren: Dao Zhou, Christoph Müssel, Ludwig Lausser, Martin Hopfensitz, Michael Kühl, Hans A. Kestler

Gene-regulatory networks control the expression of genes and therefore the phenotype of cells. Modeling and simulation of such networks can provide deep insights into the functioning of cells. Boolean networks are a commonly used technique to model gene-regulatory networks. We introduce methods to construct Boolean networks from literature knowledge and to analyze their dynamics. In particular, methods to identify and analyze attractors are presented. In simulations on three biological networks, we analyze the robustness of attractors. These evaluations confirm the biological relevance of previously identified attractors.

PDF

Autoren: J. Hanika, H.P.A. Lensch, A. Keller

We introduce a ray tracing architecture which is able to handle highly complex geometry modeled by the classic production approach of surface patches tessellated to micro-polygons, where the number of micro-polygons can exceed the available memory. Two novel techniques allow us to carry out global illumination computations in such scenes and to trace the resulting incoherent sets of rays efficiently. For one, we introduce a technique for building the BVH over tessellated patches in time linear in the number of micro-polygons. Second, we present a two-stage ray tracing system which is highly parallel and minimizes the number of tessellation steps by reordering rays. The technique can accelerate rendering scenes of billions of micro-polygons and objects with complex reflection shaders using deferred shading.

PDF

Autoren: Stephan Buchwald, Thomas Bauer, Manfred Reichert

Häufig genannte Ziele einer Service-orientierten Architektur (SOA) sind die bessere Unterstützung und Anpassbarkeit von Geschäftsprozessen sowie das Business-IT-Alignment. Diese Ziele werden heute nicht erreicht, da die bei der Implementierung eines Fachprozesses notwendigen komplexen Transformationen in einen ausführbaren Workflow schwer nachvollziehbar sind. Dadurch gehen fachliche Anforderungen verloren und es entsteht ein hoher Aufwand bei späteren Prozessanpassungen. Dieser Beitrag führt die Ebene des Systemmodells zwischen Fachbereich (Fachmodelle) und IT-Bereich (ausführbare Modelle) ein, um Geschäftsprozess-Transformationen besser zu unterstützen und den Ergebnisprozess mit den Fachbereichen abstimmen zu können. Im vorgestellten Ansatz wird ein sog. Abbildungsmodell eingeführt, das Zugehörigkeiten von Aktivitäten des Fachmodells zu denen des Systemmodells (d.h. technische Spezifikation des Informationssystems) explizit dokumentiert. Dadurch werden im Software-Entwicklungsprozess automatisierte Konsistenzprüfungen zwischen den Modellebenen möglich. Werden später Prozessanpassungen erforderlich, lassen sich die zu einer fachlichen Aktivität gehörenden technischen Aktivitäten unmittelbar erkennen, was die Durchführung der Anpassung erleichtert. Ein wesentlicher Vorteil unseres Ansatzes besteht darin, dass die Erstellung des Abbildungsmodells nur einen minimalen Aufwand verursacht, da keine komplexen Regeln, sondern nur einfache Beziehungen definiert werden müssen. Der Ansatz ist damit in der Praxis gut verwendbar, was durch eine prototypische Umsetzung unterstrichen wird.

PDF