Nenad Ivezic

An Architecture for Semantic Enterprise Application Integration Standards

Large, industry-wide interoperability projects use syntax-based standards approaches to accomplish interoperable data exchange among enterprise applications. We are investigating Semantic Web to advance these approaches. In this paper, we describe an architecture for Semantic Enterprise Application Integration Standards as a basis for experimental assessment of the Semantic Web technologies to enhance these standards approaches. The architecture relies on automated translation of the XML Schema-based representation of business document content models into an OWL-based ontology. Based on this architecture, we use Semantic Web representation and reasoning mechanisms to support consistency checking of ontological constructs and constraints specified within the ontology. The proposed architecture is relevant (1) when managing multiple enterprise ontologies derived from, and dependent on, a common ontology and (2) when dealing with model-driven integration using customizable interface models and testing of such integration efforts.

Semantic mediation for standard-based B2B interoperability

A semantic-mediation architecture advances traditional approaches for standard-based business-to-business interoperability. The architecture formally models a business domain in a reference ontology and annotates domain message schemas to define public and proprietary reconciliation rule sets. Enterprises can use the rule sets to implement standard-based message interfaces and to translate message content between their proprietary message forms. An implementation of the semantic-mediation architecture augments a general applications-integration toolset developed for the Athena European FP 6 project. The implementation demonstrates the architectures feasibility and suggests directions for future tool enhancements.

Neural network-based resistance spot welding control and quality prediction

This paper describes the development and evaluation of neural network-based systems for industrial resistance spot welding process control and weld quality assessment. The developed systems utilize recurrent neural networks for process control and both recurrent networks and static networks for quality prediction. The first section describes a system capable of both welding process control and real-time weld quality assessment. The second describes the development and evaluation of a static neural network-based weld quality assessment system that relied on experimental design to limit the influence of environmental variability. Relevant data analysis methods are also discussed. The weld classifier resulting from the analysis successfully balances predictive power and simplicity of interpretation. The results presented for both systems demonstrate clearly that neural networks can be employed to address two significant problems common to the resistance spot welding industry, control of the process itself, and nondestructive determination of resulting weld quality.

Multiagent framework for lean manufacturing

We have developed the manufacturing agent-based emulation system as an open framework for design and analysis of discrete manufacturing systems. MABES currently supports the transition from traditional to lean manufacturing in two major functions: analysis of alternative agent-based scheduling and control approaches that can be implemented across the extended enterprise; and real-time collaboration of design teams during manufacturing line design and analysis stages. MABES bases its support for these functions on two system paradigms: distributed agents and synchronous collaboration.

A neural network-based machine learning approach for supporting synthesis

The goal of machine learning for artifact synthesis is the acquisition of the relationships among form, function, and behavior properties that can be used to determine more directly form attributes that satisfy design requirements. The proposed approach to synthesis knowledge acquisition and use (SKAU) described in this paper, called NETSYN, creates a function to estimate the probability of each possible value of each design property being used in a given design context. NETSYN uses a connectionist learning approach to acquire and represent this probability estimation function and exhibits good performance when tested on an artificial design problem. This paper presents the NETSYN approach for SKAU, a preliminary test of its capability, and a discussion of issues that need to be addressed in future work.

Toward Self-Integrating Software Applications for Supply Chain Management

Each paper in this special issue deals with a particular aspect of supply chain management. Nevertheless, they all point to a common need for meaningful and timely information exchange across the supply chain. The Internet and the existing Web technologies have made the exchange of information essentially free and instantaneous. On the other hand, determining the meaning of that information, which we call integration, is still very costly. Millions of dollars and hundreds of man-years have been spent developing and coding interface specifications and software applications to achieve this integration. While this approach has been successful in the past, it is not a viable approach for the future in which the Semantic Web is becoming an important business strategy. In this paper, we discuss a new approach, called self-integration in which software applications are imbedded in an environment that allows them to integrate automatically. We first provide some background information on integration and then focus on our research on semantic querying, semantic mapping, and semantic inferencing.

Machine learning for simulation-based support of early collaborative design

The research and development of a simulation-based decision support system (SB-DSS) capable of assisting early collaborative design processes is presented. The requirements for such a system are included. Existing collaborative DSSs are shown to lack the capability to manipulate complex simulation-based relationships. On the other hand, advances within the machine learning in design community are shown to have a potential for providing, but have not yet addressed, simulation-based support for collaborative design processes. The developed SB-DSS is described in terms of its four principal components. First, the behavior-evaluation (BE) model is used to both structure individual, domain-specific decision models and organize these models into a collaborative decision model. Second, a probabilistic framework for the BE model enables management of the uncertainty inherent in learning and using simulation-based knowledge. Significantly, this framework provides a constraint satisfaction environment in which simulation-based knowledge is used. Third, a statistical neural network approach is used to capture simulation-based knowledge and build the probabilistic behavior models based on this knowledge. Fourth, since a probability distribution theory does not exist for the nonlinear neural network approaches, Monte Carlo simulation is introduced as a method to sample the trained neural networks and approximate the likelihoods of design variable values. Consequently, constraint satisfaction problem-solving capability is obtained. In addition, a mapping of the SB-DSS architecture onto a collaborative design agent framework is provided. Experimental evaluation of a prototype SB-DSS system is summarized, and performance of the SB-DSS with respect to search and usability metrics is documented. Initial results in developing the simulation-based support for collaborative design are encouraging. Lastly, a categorization of the machine learning approach and a critique of the proposed categorization scheme is presented.

A semantic-mediation architecture for interoperable supply-chain applications

This paper presents a semantic-mediation architecture that enables standards-based interoperability between heterogeneous supply-chain applications. The architecture was implemented using a state-of-the-art semantic-mediation toolset for design-time and run-time integration tasks. The design-time tools supported a domain ontology definition, message annotations, message schema transformations and reconciliation rules specifications. The run-time tools performed exchanges, transformations, and reconciliations of the messages. The architecture supports a supply-chain integration scenario where heterogeneous automotive manufacturing supply-chain applications exchange inventory information.

Enabling Semantic Mediation for Business Applications: XML-RDF, RDF-XML and XSD-RDFS transformations

In this paper, we describe results of our work in developing transformation tools that enable use of RDF-based Semantic Mediation tools for integration of business applications that have implemented XML Schema-based interfaces. Specifically, we are concerned with validating advanced Semantic Mediation solutions that required XML to RDF, RDF to XML, and the XML Schema to the RDF Schema transformations to be used by these business applications. To use the advanced integration solutions we developed all three of these transformations. We discuss the requirements for the transformation tool posed by the Semantic Mediation approach in the context of a business-to-business scenario. We analyze the related work and various related approaches. We describe our implemented transformation approach and explain the intended use of the resulting tool within a typical usage scenario.

Semantic resolution for e-commerce

We describe a research project on resolving semantic differences for multi-agent systems (MAS) in electronic commerce. The approach can be characterized as follows: (1) agents in a MAS may have their own specific ontologies defined on top of a shared base ontology; (2) concepts in these ontologies are represented as frame-like structures based on DAML+OIL language; (3) the semantic differences between agents are resolved at runtime through inter-agent communication; and (4) the resolution is viewed as an abductive inference process, and thus necessarily involves approximate reasoning.