Workshop on

Knowledge-Based Systems for Model-Based Engineering

22.08.2000

Accepted papers

This page contains abstracts and PDF files of all accepted workshop papers.

Paper 1

Knowledge Engineering in Software Product Lines

Michael Schlick and Andreas Hein

Abstract A software product-line is a collection of products sharing a common set of features that address the specific needs of a given business area [1]. The PRAISE project [2], partly founded by the European Commission under ESPRIT contract 28651 and pursued by Thomson-CSF/LCR (France), Robert Bosch GmbH (Germany), and the European Software Institute (Spain), has investigated product-line realisation and its assessment in industrial settings. A part of the project was dedicated to the validation and consolidation of proposed product-line technologies in real-scale industrial experiments. This paper presents an extract of the experimental results found by Bosch. The Bosch experiment has been located in the Car Periphery Supervision (CPS) domain. The focus during analysis was on feasibility of variability modelling with FODA [3]. The experiment has shown that the FODA model does not provide the necessary expressiveness to represent the different types of crosslinks that are obligatory to describe the domain. Therefore an extension was made to overcome this drawback. Moreover, it became clear that a lot of issues concerning the configuration of FODA models are far from being applicable. Here a solid theoretical foundation is needed first. This paper presents some basic findings.

Download full paper: PDF

Paper 2

System Design for Reusability – Task, Current State, and Future Research

Hauke Arndt , Frank Feldkamp, Michael Heinrich, Klaus Dieter Meyer-Gramann

Abstract In this paper, an IT support for vendors of transportation systems and their subsystems in the early phases of the product design is presented: SDR, System Design for Reusability. Water pumps for cooling combustion engines serve as a pilot application, an editor for the water pump designer was implemented. SDR supports a reusable and modular product design by different modeling features: systems with ports, a taxonomy with system types, interfaces with types, and modular constraints between parameters of systems. SDR stresses the principles of modular system design. This requires a thorough and consistent segmentation of the system at all levels. Moreover, a well-chosen set of interfaces between the modules is defined.
SDR serves as a design assistant. A block diagram editor allows to describe systems, ports and interfaces. Constraint propagation and constraint solving is seamlessly integrated in the modeling of the system structure. SDR supports a least commitment approach during the product design process in order to capture uncertainty in early design phases adequately and to support reusability. Future research will tackle specific issues of reasoning, user interactions, and documentation.

Download full paper: PDF

Paper 3

Generating Fault Trees from Mixed Quantitative and Qualitative Electrical Device Models

Heiko Milde and Lothar Hotz

Abstract Computer diagnosis systems grounded on hand-crafted fault trees are wide-spread in industrial practice. Since the complexity of technical systems increases and innovation cycles get shorter, the need for systematic fault tree generation and maintenance arises. In this paper, the MAD system is introduced which generates fault trees based on models of technical devices. In addition to qualitative device modeling, MAD allows context-dependent quantitative measurement modeling to secure accurate fault identification. Due to quantitative measurement models, tests in MAD's fault trees concern quantitative parameter threshold values which corresponds to usual industrial practice. Hence, context-dependent quantitative measurement modeling is essential for MAD's industrial acceptance. We have successfully evaluated the MAD system in cooperation with the German forklift manufacturer STILL GmbH Hamburg.
Keywords: Fault tree generation, diagnostic decision tree generation, model-based-diagnosis, qualitative modeling, mixed modeling

Download full paper: PDF

Paper 4

Towards Model-based Engineering: Failure Analysis with MDS

Jakob Mauss, Volker May, Mugur Tatar

Abstract Model-based engineering supports different engineering tasks using rich digital models of products. Such models should improve product-related communication between engineers, should increase knowledge sharing and the re-use of partial solutions among different products and among different engineering tasks, and should offer new chances for systematic product design and validation. In this paper we present MDS, a tool that implements model-based engineering for various tasks related to the analysis of the behavior of a product when failures are present. We describe the modelling framework and the model analysis services provided by MDS, some recent applications of MDS, and some lessons learned from our work with applications.

Download full paper: PDF

Paper 5

Model-based Diagnosis and Manufacturing Control

Martin Ilkerl, Markus Stumptner, Franz Wotawa

Abstract Given the increasing use of digital control systems in manufacturing plants, the theoretical potential for error monitoring and diagnosis has significantly increased. However, conventional monitoring systems often still suffer from the problems of past decades where automated error recognition is nonexistent and the monitoring system restricts itself mainly to communications and signal processing. As a result, the diagnostic load in case of malfunctions is left to the operator. We describe a model-based diagnosis system that is integrated into the control of a manufacturing plant by providing an additional layer atop the existing monitoring system. The diagnosis system provides flexible, high-level and resilient diagnosis capability and can directly use the monitoring system’s reports as observations for the diagnosis process. We also discuss general principles of developing models for this domain and show a simple model used in our current diagnosis/monitoring prototype system.

Download full paper: PDF

Paper 6

An Environment and Language for Industrial Use of Model-based Diagnosis

Gerhard Fleischanderl, Herwig Schreiner, Thomas Havelka, Markus Stumptner, Franz Wotawa

Abstract Model-based diagnosis provides a well founded theory and a set of algorithms for finding and fixing a misbehavior caused by components. Actually applying model-based diagnosis effectively requires a flexible implementation which is capable of handling the differing requirements of multiple application domains. The diagnosis framework described in this paper has been developed for the purpose of being used in an industrial setting. It derives a significant part of its effectiveness from being integrated with a component oriented language for describing diagnosis models. The framework itself contains a class library comprising several different diagnosis engines having a standardized interface and allows rapid prototyping of diagnosis applications. The paper describes the framework, shows the engineering application domains where the framework was applied, and gives an overview of the capability of the system description language AD2L.

Download full paper: PDF

Paper 7

Using Search in Knowledge Based Engineering

Andreas Junghanns, Rüdiger Klein

Abstract Search has been instrumental in many well known successes of AI problem solving. These successes are currently mostly restricted to well structured domains like games or scheduling. The resulting search spaces can be searched using special purpose search heuristics and strategies. In this position paper we formulate the problems and challenges encountered when these successes are to be extended to more complex domains. Engineering provides a good test bed for these kinds of efforts: though quite complex and diverse, engineering applications have a clear structure and are (generally) well understood. Many different problem solving methods are typically used in combination. The resulting search spaces tend to be huge, they are typically dynamic, and include continuous as well as discrete dimensions. We analyze in which way current search techniques can be extended and modified in order to be applicable under these circumstances.

Download full paper: PDF

Paper 8

Applying a generic constraint solving technique to engineering design

Hiroyuki Sawada, Xiu-Tian Yan

Abstract As a design solution becomes more concrete and detailed during engineering design, more design parameters are introduced to specify the solution quantitatively. This increased number of design parameters causes challenges and difficulties to a designer in terms of gaining an insight into the significance of these design parameters and their influence on the overall performance of a product. This in turn can easily lead a designer to generating a less optimal final design solution. In order to overcome such difficulties, a design support system based on a generic constraint solving technique is derived to support engineering design. The system is concerned with three types of constraints, namely kinematic, energetic and spatial constraints of a product. Standard components can be selected from the system component library and a product can be subsequently configured using these components. The system then can automatically generate all the related constraints and a design solution are quantified with solution of constraint problem solving. The system can also solve an incompletely defined design solution.

Download full paper: PDF

Paper 9

AutoSteve: Automated Electrical Design Analysis

Chris Price

Abstract AutoSteve performs automated electrical design analysis based on qualitative simulation and functional abstraction. It is the first commercial product capable of performing these tasks for complex automotive systems. It has been deployed at automotive manufacturers for several years, and produces FMEA and sneak circuit analysis reports much more quickly and consistently than they could be produced without its assistance. It is an extended version of a paper presented in PAIS-2000.

Download full paper: PDF

Paper 10

Knowledge Models in Engineering Design

Michael Valášek, Zdenek Zdráhal

Abstract The paper deals with the description of knowledge models used by the authors for solving different problems in engineering design, namely the problem of extending traditional computational tools CAE wihin configuration design by additional procedures of knowledge routine processing and the problem of analysis of engineering design procedures in order to transform them from traditional into concurrent engineering design.

Download full paper:PDF

Paper 11

Design as a Problem of Requirements Explication

Martin Dzbor

Abstract Engineering design is a human activity using different knowledge sources. From the point of availability it is possible to distinguish well-structured, explicit knowledge as opposed to tacit, implicit and often experience-based knowledge. Each type plays a particular role in engineering design, and thus in knowledge-based design support systems (KBDSS). This paper addresses several issues with KBDSS. It begins with a discussion of the processes underlying engineering design. Further, it presents a generic model of engineering design and shows the role of reflection in design. Based on such a framework, implications for the design support are drawn and justified using a well-known example.

Download full paper: PDF

This page is maintained by mst@dbai.tuwien.ac.at and was last modified on 14/06/00.