Semantic SOA 4 SOA Supply Networks Leonid Sheremetov*, Christian Sanchez-Sanchez** *St. Petersburg Institute for Informatics and Automation of the Russian Academy of Science 14 line 39, St. Petersburg, Russia (Tel: 812-328-8071; e-mail: lsheremetov@mail.ru). ** CINVESTAV-IPN, Av. IPN No. 2508 CP 07360. Mexico, (e-mail: christiansanchezs@gmail.com) Abstract: In a complicated business network finding a supplier can be a very time consuming task. In advanced supply networks like Build-to-Order supply chains, this task should be carried out under time constraints and under uncertainties both in suppliers and in the orders. The technology of semantic service oriented architecture is aimed to support such kind of tasks, enabling construction of self- organizing adaptive supply chain networks. A novel approach to network members’ discovery and selection based on their profiles is described. The approach is grounded in a method for service discovery with incomplete information using query expansion techniques. The usage of the approach is illustrated by example. Keywords: supply network, semantic SOA, service discovery. 1. INTRODUCTION Nowadays agility, reactivity, flexibility and adaptivity of a supply chain play a key role for the success of an enterprise in gaining competitiveness [SAP, 2002]. Complexity theories have portrayed the supply chain as a complex adaptive system whose behaviour is influenced by self organizing mechanisms. A Self-Organizing Adaptive Supply Chain Network (SOA-SCN) is an interconnected network of multiple entities (or agents) that exhibit adaptive action in response to changes in both the environment and the system of entities itself [Choi et al., 2001]. A good example of such adaptivity is Build-to-Order (BTO) supply chains strategy, when customer orders are introduced prior to production, so that production channels have to be dynamically configured under demand [Kathawala and Wilgen 2005; Gunasekarana and Ngaib, 2005]. In order to efficiently and precisely configure the channels, semantically enabled technologies permitting aggregation of relevant knowledge from heterogeneous knowledge sources are required. The importance of knowledge sharing in flexible supply networks both at technical and semantic levels has been shown by Sandkuhl et al. (2007) and Smirnov et al. (2008). At the technical level the interoperability is addressed by SOA, at the semantic level – by ontologies. The approach presented in this paper relies on these levels’ integration in Semantic Service Oriented Architectures (SOA). The advantages of semantic SOA for enterprise application integration were discussed for instance in Gou et al. (2008) and Contreras and Sheremetov (2008). In this paper we show how semantic SOA could assist in automation of a dynamic generation of on-demand supply chain channels of a BTO supply chain. This is one of the tasks of a Competence Management System [Smirnov et al., 2008]. The key to achieve the automation relies mainly on solutions of three issues: (1) how to make services of supply chain interoperable both syntactically and semantically; (2) how to automatically discover, based on the syntactic and semantic descriptions, the most appropriate information and services; and (3) assemble them to build the composite service. In this paper we describe an approach to solving these problems inspired in service discovery method using query expansion techniques to facilitate matchmaking and service discovery proposed by Sanchez and Sheremetov (2008). The rest of the paper is organized as follows. In the following section the baseline technologies are briefly analysed and the conceptual model of the framework is defined. In Section 3, basic definitions for service similarity and matching are introduced along with the algorithm for dynamic service composition. In Section 4 an example illustrating the proposed approach is presented. Finally, the concluding remarks summarize the obtained results. 2. BACKGROUND We argue that in addition to the traditional and well- understood supply chain management solutions of planning, execution, coordination, and networking, emerging technologies of Semantic SOA – agents, ontology, and Web services – are equally significant for SCN’s self-organization and adaptation (SOA 4 SOA paradigm). Instead of responding to requests for information, they permit intelligently anticipate, adapt, and automate SCN activities. 2.1 Supporting technologies The forming of on-demand supply chain channels is dynamic and uncertain in nature due to the frequent changes of demand and availability of suppliers [Lee and Tang, 1997]. In order to exploit the benefits of available-to-promise Proceedings of the 13th IFAC Symposium on Information Control Problems in Manufacturing Moscow, Russia, June 3-5, 2009 978-3-902661-43-2/09/$20.00 © 2009 IFAC 1304 10.3182/20090603-3-RU-2001.0445