This article gives an overview of AI (Artificial Intelligence) plan-ning techniques and discusses their application to the Web service composition problem.

As the number of available Web services increases there is a growing demand to realise complex business processes by combining and reusing available Web services. The reuse and combination of services results in a composition of Web services that may also involve services provided in the Internet. With semantically described Web services, an automated matchmaking of capabilities can help identify suitable services. To address the need for semantically defined Web services, OWL-S and WSML have been proposed as competing semantic Web service languages. We show how the proposed semantic Web service languages can be utilized within a model-driven methodology for building composite Web services. In addition we combine the semantic-based discovery with the support for processing QoS requirements to apply a ranking or a selection of the candidates. The methodology describes a process which guides the developer through four phases, starting with the initial modelling, and ending with a new composite service that can be deployed and published to be consumed by other users. 1

Abstract. The description of interactions among Web Services with regard to the exchange of messages, their composition, and the sequences in which they are transmitted and received is an important problem. While some efforts tackle this problem by defining standards for static processes others aim at composing complex services on-the-fly using AItechniques. The present paper follows this AI approach and suggests a solution that uses Answer Set Programming (ASP) as a mean to achieve the composition goal. In particular, it shows how Web service descriptions and customer requests are mapped to the input language of the ASP software DLV 1 [1]. Furthermore, it shows how composition goals and constraints can be defined to guide the composition challenge. Finally, it presents a tool that serves as a framework to build complex services and that uses DLV as one of its back-end solvers. The approach was evaluated in the first Web service composition contest and has won the first prize 2. 1 Web Service Composition In recent research related to Web services, several initiatives are aiming to provide means to allow easy integration of heterogeneous systems by defining languages, methods, or platforms. Some of these efforts (e.g. BPEL [2], OWL-S [3]) aim towards static processes with a priori knowledge of all possible execution paths, or by applying AI methods to compose complex services on-the-fly for an individual service requester. A Web service is a collection of protocols and standards used for exchanging data between applications. Software applications written in various programming languages and running on various platforms can use web services to exchange data over computer networks like the Internet in a manner similar to interprocess communication on a single computer. 1 DLV is an international cooperation between the University of Calabria and

Abstract not found

RESUME: Cet article présente les résultats d'une étude de simulation d’un système flexible de production (FMS) qui se compose de sept machines, une station de chargement, une station de déchargement, et six types de pièces différentes. En raison de l'existence de centres d'usinage identiques dans le système, les types de pièces ont des routages alternatifs. De plus, au niveau de chaque machine, FIFO (First In First Out) est employé comme règle de séquencement. L’objectif de cette étude est d’analyser l'impact du ré-ordonnancement en temps réel des pièces contenues dans la station de chargement sur les performances du système lorsque les métaheuristiques suivantes (les colonies de fourmis, les algorithmes génétiques, le recuit simulé, la recherche tabou, les essaims particulaires et l’électromagnétisme) sont appliquées pour résoudre le problème de sélection de routages alternatifs en temps réel. Les résultats obtenus ont montré que pour chaque métaheuristique que nous avons étudiée, le ré-ordonnancement améliore les performances pour le taux de production, le taux d’utilisation des machines et du système de transport pour un système flexible de production saturé, même avec la présence de pannes. MOTS-CLES: Système flexible de production, routage alternative, règle de routage alternative, métaheuristique, réordonnancement 1

One main characteristic of virtual organizations are short-term collaborations between business partners to provide efficient and individualized services to customers. The MOVE project targets at a methodology and a software framework to support such flexible collaborations based on process oriented design and communication by Web services. MOVEs framework supports the graphical design and verification of business processes, the execution and supervision of processes in transaction-oriented environment, and the dynamic composition and optimization of processes. A business process may be composed from a set of Web services, deployed itself as Web service and executed in the framework. The composition of processes from Web services is implemented with methods from AI-planning. We map Web service descriptions, domain knowledge and customer requests into the input language of AI-planners. We show the performance of our program and give some implementation details. Finally we conclude with some remarks about further research. 1.