2434 Semantic Knowledge Transparency in E-Business Processes managing the integration of value chain activities over distributed and heterogeneous information platforms such as the Internet, is a challenging WDVN ZLWK ODUJH SRWHQWLDO EHQH¿WV $OWKRXJK technical integration of systems is essential, a FRPPRQODQJXDJHWRH[SUHVVFRQWH[WVSHFL¿F constructs and relevant business rules to assist autonomous system entities and decision makers WRVROYHVSHFL¿FEXVLQHVVSUREOHPVLVHVVHQWLDO (Stal, 2002). Disparate technical systems need the ability to share data, information, and knowl- edge. A common and shared understanding of WKH GRPDLQVSHFL¿F FRQFHSWV DQG WKH UHODWLRQV between them is critical for creating integrative views of information and knowledge in e-business processes. However, there is paucity in research on distributed information and knowledge shar- ing that provides a unifying process perspective to share information and knowledge (Oh & Park, 2003) in a seamless manner. The Semantic Web is a key component for realizing the vision of semantic knowledge trans- parency in e-business processes. The Semantic Web provides the technical foundations to sup- SRUWWKHWUDQVSDUHQWÀRZRIVHPDQWLFNQRZOHGJH representation to automate, enhance, and coordi- nate collaborative inter-organizational e-business processes (Singh, Iyer, et al., 2005). The Semantic Web vision comprises ontologies for common semantics of representation and ways to interpret ontology; knowledge representation for structured collections of information and inference rules for automated reasoning in a single system; and intelligent agent to collect content from diverse sources and exchange data enriched with seman- tics (Berners-Lee, Hendler, & Lassila, 2001). This vision provides the foundation for the semantic framework proposed in this research. This chapter is structured as follows. First, we conduct a review and analysis of the relevant literature in the areas of e-business, KM, and the Semantic Web. Second, the conceptualization of the e-business process universe of discourse and its description logic are developed. Third, we use an intelligent infomediary-based e-marketplace as a scenario to illustrate how semantic knowledge transparency can be used to achieve the coordi- nation of activities and resources across inter- organizational systems. Finally, future research issues and conclusions are stated. BACKGROUND Based on existing research in e-business, KM and the Semantic Web, an innovative approach to achieve semantic knowledge transparency is developed. We use a process perspective to integrate knowledge of resources involved in a process and process knowledge including process PRGHOVDQGZRUNÀRZVXVHGLQSURFHVVDXWRPDWLRQ In order to achieve semantic knowledge transpar- ency, we develop theoretical conceptualizations using ontological analysis that will be formalized using DLs. The ontology will support a common vocabulary for transparent knowledge exchange among inter-organizational systems of business p a r t n e r s o f a v a l u e c h a i n , s o t h a t s e m a n t i c i n t e r o p - erability can be achieved. The foundations of the proposed approach are conceptually represented in Figure 1 and explanations follow. E-Business Electronic data interchange (EDI) is an informa- tion technology that allows business partners to send and receive commercial documents in an electronic format (Hansen & Hill, 1989). Under EDI proprietary value-added networks data dis- closure and information transparency were not a concern. Interestingly, EDI by itself does not provide market transparency (Zhu, 2004). Nowadays, businesses are moving from EDI WR:HEEDVHGV\VWHPV,QIDFWPDQ\¿UPVKDYH adopted e-business models to improve their collaborative capabilities (Segars & Chatterjee, 2003). Regarding business processes, they are typically modeled as deterministic, action-event 2435 Semantic Knowledge Transparency in E-Business Processes V HTX H QF H VL Q ZR UN ÀR Z ED V H GL Q IRU P D W LR QV \ VW H P V DQG ZRUNÀRZ DXWRPDWLRQ V\VWHPV :RUNÀRZV establish the logical order of execution between individual task units that comprise intra-or- ganizational and inter-organizational business SURF HVVHV7 KH:RUN ÀRZ0D QDJH PHQW&RDO LW LRQ (www.wfmc.org) describes business process as ³DVHTXHQFHRIDFWLYLWLHVZLWKGLVWLQFWLQSXWVDQG outputs and serves a meaningful purpose within an organization or between organizations” (Dust- dar, 2004 p. 460). $ SURFHVV GH¿QLWLRQ LV WKH UHSUHVHQWDWLRQ of a business process in a form which supports automated manipulation, such as modeling, or HQDFWPHQW E\D ZRUNÀRZPDQDJHPHQW V\VWHP :I06 7KH SURFHVV GH¿QLWLRQ FRQVLVWV RI D network of activities and their relationships, cri- teria to indicate the start and the termination of the process, and information about the individual activities, including participants and data (www. wfmc.org). Business processes can thus be gener- DO L]HG DVKDY LQJD³ EHJL Q´DQGDQ³HQG´ SRL QWD QG a series of intermediate tasks that are performed in sequence on some entity, object, or activity. In its simplest case, an e-business process may KDYH HDFKZRUNÀRZ DFWLYLW\ SHUIRUPHG ZLWKLQ a single organization; while in the most general and extensible case, each individual activity may be performed by a different partner organization. 0RVWLQWHURUJDQL]DWLRQDOZRUNÀRZVZRXOGIDOO somewhere in between these end points. Singh, Iyer, et al. (2005) explain that e-business processes require transparent information and semantic knowledge transparency among business partners. The consequent lack of transparency in L Q IR U P DWL RQ ÀRZ D F UR V VW K HYD OX HFK DL Q F RQ W L QX H V to hinder productive and collaborative partner- VKLSDPRQJ¿UPVLQe-marketplaces. Moreover, the lack of transparency in business-to-business (B2B) e-marketplaces increases the uncertainty and perceived risks and hampers trusted relation- ships among business partners. E-Marketplace The main roles of e-marketplace are: (1) discovery – of buyers and suppliers that meet each other’s requirements; (2) facilitation – of transactions to HQDEOHLQIRUPDWLRQÀRZVOHDGLQJWRWKHÀRZRI good and services among buyers and suppliers; and (3) support – of decision process leading to the development of collaborative relationships Figure 1. Conceptual representation of semantic knowledge transparency and integration. Semantic Web Ontology, Knowledge Representation, Intelligent Agents e-Business e-Marketplace , Infomediary Knowledge Management Organizational Knowledge, Interorganizational Processes Coordination Semantic Knowledge Transparency Semantic Web Ontology, Knowledge Representation, Intelligent Agents e-Business e-Marketplace , Infomediary Knowledge Management Organizational Knowledge, Interorganizational Processes Coordination Semantic Knowledge Transparency 2436 Semantic Knowledge Transparency in E-Business Processes between e-marketplace participants (Bakos, 1998). The value added to the process by the e- marketplace is in providing information to buyers and suppliers about each others’ capabilities and requirements. E-marketplace is a mechanism to VWUHDPOLQHLQIRUPDWLRQÀRZLQVXSSO\FKDLQDQG re-balance the information asymmetry (Zhu, 2002). E-marketplaces offer value-added ser- YLFHVE\OHYHUDJLQJLQGXVWU\VSHFL¿FNQRZOHGJH through deciphering complex information and contribute to transaction cost reduction. How- ever, the lack of integration of information and knowledge across the e-value chain continues to hinder productive and collaborative partnerships DPRQJ¿UPVLQHPDUNHWSODFHV Infomediary In e-marketplaces a new kind of intermediaries has emerged: Infomediary. Grover and Teng (2001) GH¿QHLQIRPHGLDU\DV³HFRPPHUFHFRPSDQLHV leveraging the [power of] the Internet to unite EX\HUVDQGVXSSOLHUVLQDVLQJOHHI¿FLHQWYLUWXDO marketspace to facilitate the consummation of a transaction” (p. 79). In this chapter, we argue that in the context of e-marketplaces, intermediaries have evolved into infomediaries that add value to their stakeholders by deciphering complex product information and matching buyers’ needs with sellers’ products and/or services. Grover and Teng focus on the critical information-providing role of the market and identify the roles played by electronic intermediaries, or infomediaries. An infomediary is an emergent business model adopted by organizations in response to the enormous increase in the volume of information available and the critical role of information in enabling processes in electronic markets. Info- mediaries perform an indispensable function by matching buyers’ needs with suppliers’ products and services to facilitate transactions. There is a wealth of market information exchanged through the infomediaries as they perform these functions. As a result, Infomediaries become vital resources of knowledge about the nature of exchanges in the e-marketplace. A n a n a l y s i s o f t h e i n f o m e d i a r y b u s i n e s s m o d e l shows that individual buyers and suppliers seek distinct goal-oriented information capabilities from the infomediary—they provide decision parameters through their individual demand or supply functions. This is essentially a discovery activity with buyers and suppliers searching for a match of their requirements through infomediar- LHV7KLVGLVFRYHU\SURFHVVLVLQÀXHQFHGE\KLV- torical information including the past experiences of other buyers’ reliability and trustworthiness of the supplier. The infomediary business model can provide valuable information to this deci- sion process through its role as the repository of experiential knowledge of transactional histories for both buyers and suppliers. This information can be used to develop knowledge that informs discovery of buyers and suppliers for subsequent transactions. A realization of the need for greater collaboration among trading partners is fueling the growth of KM to help identify integrative and interrelated elements to enable collaborations. Knowledge Management .0FDQEHGH¿QHGDV³DSURFHVVWKDWKHOSVRU- JDQL]DWLRQV ¿QG VHOHFW RUJDQL]H GLVVHPLQDWH and transfer important information and expertise necessary for activities such as problem solving, dynamic learning, strategic planning, and deci- sion making” (Gupta, Iyer, & Aronson, 2000, S.0LQFOXGLQJWKHFRGL¿FDWLRQVWRUDJH retrieval, and sharing of knowledge, transpires in WKHFRQWH[WRIDSURFHVVVFLHQWL¿FJRYHUQPHQWDO or commercial. Explicit knowledge, declarative enough to be represented using standards-based knowledge representation (KR) languages allows for knowledge to be interpreted by software and shared using automated reasoning mechanisms to reach useful inferences. While all knowledge cannot be explicated and be effectively represented and reasoned with using decidable and complete 2437 Semantic Knowledge Transparency in E-Business Processes computational techniques; it is useful to focus on explicit, declarative KR using computationally feasible KR languages to build effective and useful NQRZOHGJHEDVHGV\VWHPV+DPHOLGHQWL¿HV that knowledge transparency is directly related to HDVHRIWUDQVIHU,QOLQHZLWKWKHQRWLRQRI¿UPVDV repositories of productive knowledge (Demsetz, 1998), where knowledge resources are primary concern, managing cooperative relationships is I U H TX H QW O\D SUR F HV V RI P D Q DJ L Q JN QR ZOH G JH ÀRZ V (Badaracco, 1991). Furthermore, transparency is critical to busi- ness partnerships, lowering transaction costs EHWZHHQ¿UPVDQGHQDEOLQJFROODERUDWLYHFRP- merce (Tapscott & Ticoll, 2003). We focus on two VSHFL¿FW\SHVRINQRZOHGJHLQWKLVUHVHDUFK 1. Component knowledge: Component knowledge includes descriptions of skills, technologies, tangible and intangible resources and is amenable to knowledge exchange (Hamel, 1991; Tallman, Jenkins, Henry, & Pinch, 2004). 2. Process knowledge: Process knowledge is typically embedded in the process models RIZRUNÀRZPDQDJHPHQWV\VWHPVRUH[LVWV as coordination knowledge among human agents to coordinate complex processes. Component and process knowledge are central to activities of human and software agents in inter-organizational e-business processes; there- fore, the standard representation of both type of knowledge is fundamental to achieve semantic knowledge transparency. Newell (1982) regards NQRZOHGJHDV³ZKDWHYHUFDQEHDVFULEHGWRDQ agent, such that its behavior can be computed according to the principle of rationality” (p. 105). 7KLVGH¿QLWLRQIRUPVDEDVLVIRUIXQFWLRQDO.0 using agents, human, and software when using explicit, declarative knowledge that is represented using standards-based knowledge representation languages that can be processed using reasoning mechanisms to reach useful inferences. Inter-Organizational Process Coordination Inter-organizational processes allow collaborating organizations to provide complementary services through networks of collaborating organizations (Dyer, 2000; Sawhney & Parikh, 2001). Here, WKHUHVRXUFHEDVHGYLHZRI¿UPVZLWK IRFXVHG capabilities is replaced by a network of organi- zations with a focal enterprise that coordinates resources of collaborating organizations to execute processes (Sawhney & Parikh, 2001). Complexities of coordinating inter-organizational processes require knowledge-driven coordination structures to determine decision authority and knowledge sources (Anand & Mendelson, 1997). The knowledge-integrated system incorporates the coordination mechanism and offers authorized resource matching in processes. Processes are decomposed into activities organized by gener- alization-specialization hierarchies and require coordination mechanisms to manage dependen- cies (Malone & Crowston, 1994). Coordination RIDFWLYLWLHVLVHPEHGGHGLQSURFHVVZRUNÀRZV and WfMS since they essentially deal with issues of task-task and task-resource dependencies and their coordination (Kishore, Sharman, Zhang, & Ramesh, 2004). Coordination constructs used in this proposed research are based on Malone, Crowston, and Herman (2003) and are similar to those in Van der Aalst and Kumar (2003). The complexity of coordinating e-business processes and the increasing demand by customers for complete solutions over single products re- quires knowledge-driven coordination to provide intelligent support to determine decision authority and knowledge sources in a value network. Al- liances are seldom forged to co-produce single products; they increasingly entail developing com- plex systems and solutions that require resources of multiple partners (Doz & Hamel, 1998). This requires integrative architecture with reasoning ability using knowledge about business processes within a value network. The integrated informa- 2438 Semantic Knowledge Transparency in E-Business Processes tion system as an integ ral par t of t he coord ination structure can offer enhanced matchmaking of resources and coordination of activities to allow the value network to respond to dynamic customer GHPDQGHI¿FLHQWO\DQGHIIHFWLYHO\$VRUJDQL]D- tions become increasingly global and distributed in nature, their reliance on inter-organizational LQIRUPDWLRQÀRZVZLWKSDUWQHURUJDQL]DWLRQVLV integral to e-business processes. Integrating knowledge resources across col- laborating organizations requires knowledge transparency for global, inter-organizational, access to knowledge resources. Here, semantic knowledge transparency refers to the dynamic RQGHPDQG DQG VHDPOHVV ÀRZ RI UHOHYDQW DQG unambiguous, machine-interpretable knowledge resources within organizations and across in- ter-organizational systems of business partners engaged in collaborative processes. A process view of knowledge integration incorporates man- agement of component knowledge and process knowledge for integrated inter-organizational systems that exhibit knowledge transparency. The effective standardizations and adaptability afforded by integrative technologies that support the transparent exchange of information and knowledge make inter-organizational e-business relationships viable. Semantic Web Another theoretical foundation of the semantic knowledge transparency in e-business processes is the concept of the Semantic Web. The Semantic Web is an extension of the current Web in which LQIRUPDWLRQLVJLYHQ³ZHOOGH¿QHGPHDQLQJ´WR DOORZPDFKLQHVWR³SURFHVVDQGXQGHUVWDQG´WKH information presented to them (Berners-Lee et al., 2001, p. 35). According to Berners-Lee, the Semantic Web comprises and requires knowledge representation, ontologies, and agents in order to function (Figure 2 shows the different layers of the Semantic Web architecture): • Knowledge representation:Structured col- lections of information and sets of inference rules that can be used to conduct automated reasoning. Knowledge representations must be linked into a single system. • Ontologies: Systems must have a way to discover common meanings for entity rep- resentations. In philosophy, ontology is a theory about the nature of existence; in sys- tems, ontology is a document that formally GHVFULEHVFODVVHVRIREMHFWVDQGGH¿QHVWKH relationship among them. In addition, we need ways to interpret ontology. • Agents: Programs that collect content from diverse sources and exchange the result with RWKHUSURJUDPV$JHQWVH[FKDQJH³GDWDHQ- riched with semantics.” Intelligent software agents can reach a shared understanding by exchanging ontologies that provide the vocabulary needed for discussion. Agents FDQHYHQ³ERRWVWUDS´ new reasoning capa- Figure 2. Semantic Web representation layers (Berners-Lee et al., 2001) 2439 Semantic Knowledge Transparency in E-Business Processes bilities when they discover new ontologies. Semantics makes it easier to take advantage of a service that only partially matches a request. (Lee et al. 2001, p. 37) Given the importance of the Semantic Web components to achieve processes integration and automation, we analyze in more detail relevant work in the areas of ontologies, DLs, and intel- ligent agents in the next two subsections. Ontologies and Description Logics Ontologies provide a shared and common un- GHUVWDQGLQJ RI VSHFL¿F GRPDLQV WKDW FDQ EH communicated between disparate application systems, and therein provide a means to integrate the knowledge used by online processes employed by organizations (Klein, Fensel, Van Harmelen, & Horrocks, 2001). Staab, Studer, Schnurr, and Sure (2001) describe an approach for ontology-based KM through the concept of knowledge metadata, which contains two distinct forms of ontologies that describe the structure of the data itself and issues related to the content of data. Jasper and Uschold (1999) identify that ontologies can be used for: (1) knowledge reuse; (2) knowledge VSHFL¿FDWLRQ FRPPRQ DFFHVV RI KHWHURJH- neous information; and (4) search mechanisms. We refer the reader to Kishore et al. (2004) for a more comprehensive discussion of ontologies and information systems. Ontology documents can be created using Foundation of Intelligent Physical Agents (FIPA)- compliant content languages like business process execution language (BPEL), resource descrip- tion framework (RDF), Web ontology language (OWL), and DARPA agent markup language (DAML) to generate standardized representations of the process knowledge. The structure of ontology documents will be based on DLs. DLs are logical formalisms for knowledge representation (Gomez-Perez, Fer- nandez-Lopez, & Corcho, 2004; Li & Horrocks, 2004). DLs are divided into two parts: (1) TBox, which contains intentional knowledge in the form of a terminology and is built through declarations that describe general properties of concepts; and (2) ABox, which contains extensional knowledge, ZKLFKLVVSHFL¿HGE\WKHLQGLYLGXDORIWKHGLV- c o u r s e d o m a i n ( B a a d e r , C a l v a n e s e , M c G u i n n e s s , Nardi, & Patel-Schneider, 2003; Gomez-Perez et al., 2004). I n t h i s s t u d y, w e a d o p t t h e S H I Q D L s p r e s e n t e d b y L i a n d H o r r o c k s ( 2 0 0 4 ) . T h e y a r g u e t h a t S H I Q’s expressive power made it to be equivalent to DAML+Ontology Inference Layer (OIL). In addi- tion, OWL is based on the SH family of descrip- tion logics which supports Boolean connectives, including intersection, union, and complements, restrictions on properties transitive relationships and relationship hierarchies. Standardized by the World Wide Web Consortium (W3C), OWL is the leading approach to Semantic Web ontologies using DL as its fundamental KR mechanism. Ontological analysis results in ontology descrip- tions that are presented formally through DL for theoretical soundness; and in machine-readable format using OWL and OWL-DL to provide practicality for the model. In addition, software reasoners, such as Racer, support concept con- sistency checking, TBox reasoning, and ABox reasoning on models developed using SHIQ-DL translated into OWL-DL. These provide the basis for semantic knowledge transparency to support the e-business processes. Intelligent Agents $QLQWHOOLJHQWDJHQWLV³DFRPSXWHUV\VWHPVLWX- ated in some environment and that is capable of ÀH[LEOHDXWRQRPRXVDFWLRQLQWKLVHQYLURQPHQW in order to meet its design objectives” (Jennings & Wooldridge, 1998, p. 8). The agent paradigm can support a range of decision-making activity including information retrieval; generation of alternatives; preference order ranking of options and alternatives; and supporting analysis of the 2440 Semantic Knowledge Transparency in E-Business Processes DOWHUQDWLYHJRDO UHODWLRQVKLSV 7KH VSHFL¿F DX- tonomous behavior expected of intelligent agents depends on the concrete application domain and the expected role and impact of intelligent agents on the potential solution for a particular problem for which the agents are designed to provide cog- nitive support. Criteria for application of agent technology require that the application domain should show natural distributivity with autono- mous entities that are geographically distributed and work with distributed data; require ÀH[LEOH interaction without a priori assignment of tasks to actors; and be embedded in a dynamic environ- ment (Muller, 1997). Papazoglou (2001) provides a complete discussion of the use of intelligent agents to support e-business. A fundamental implication is that knowledge must be available in formats that allow for process- ing by software agents. Intelligent agents can be u s e d f o r K M t o s u p p o r t s e m a n t i c e - b u s i n e s s a c t i v i - ties. The agent abstraction is created by extending an object with additional features for encapsula- tion and exchange of knowledge between agents to allow agents to deliver knowledge to users and support decision-making activity (Shoham, 1993). Agents work on a distributed platform and enable the transfer of knowledge by exposing their public methods as Web services using Simple Object Access Protocol (SOAP) and Extensible Markup L a n g u a g e ( X M L). I n t h i s r e s p e c t , t h e i n t e r a c t i o n s among the agents are modeled as collaborative interactions, where the agents in the multi-agent community work together to provide decision support and knowledge-based explanations of the decision problem domain to the user. A recent extension of the Semantic Web is the vision of semantic e-business. Singh, Iyer, et al. GH¿QHVHPDQWLFHEXVLQHVV DV³DQDSSURDFK to managing knowledge for coordination of e-busi- ness processes through the systematic application of Semantic Web Technologies” (p. 20). Semantic e-business leverages Semantic Web technologies DQGFRQFHSWVWRVXSSRUWWKHWUDQVSDUHQWÀRZRI semantically enriched information and knowledge and enables collaborative e-business processes within and across organizational boundaries. In addition, the Semantic Web aids intelligent agents to organize, store, retrieve, search, and match information and knowledge for effective collaboration among semantic e-business par- ticipants. It has been recognized that candidates for applications of semantic e-business include supply chain management and e-marketplaces (Singh, Iyer, et al., 2005). In this study, we apply the vision of semantic e-business in conjunction with the other theoretical foundations to explain how semantic knowledge transparency can be achieved in the context of intelligent, infomedi- ary-based e-marketplace. CONCEPTUALIZATION OF THE E-BUSINESS PROCESS UNIVERSE OF DISCOURSE 2QWRORJ\ UHSUHVHQWV VWUXFWXUHG DQG FRGL¿HG knowledge of the conceptualizations, including concepts, relationships, and constraints, for a domain of interest (Kishore et al., 2004). Fox, Barbuceanu, Gruninger, and Lin (1998) explain that organizations are a set of constraints on the activities performed by organizational agents, which can play one or more roles. At the same time, each role is designed with a set of goals and authorization levels that allow the agent to DFKLHYH WKH SUHGH¿QHG JRDOV ,Q DQ HEXVLQHVV process, a human or software agent represents a business enterprise and performs activities on its behalf. Agents perform the individual business activities that comprise the e-business process. Business activities require access to resources of the organization in order to perform the e-busi- ness process. Activities are operations performed by agents on individual resources owned by a business enterprise. Resources, owned by vari- ous owner organizations or business enterprises, coordinate activities that are performed on them. In the e-business process universe of discourse, 2441 Semantic Knowledge Transparency in E-Business Processes information and knowledge are central resources. They are used by actors in business enterprises to perform their assigned tasks (activities) in order to accomplish their goals. In this chapter, we XWLOL]HDSUDJPDWLFGH¿QLWLRQRINQRZOHGJHWKDWLV explicit and declarative enough to be represented by a standards-based knowledge representation language or formalism. Additionally, we constrain this declarative knowledge as amenable to being processed through some reasoning mechanism to reach useful inference. The essential set of concepts fundamental to model e-business processes are: business enterprise, agent, business activity, resource, coordination, information, and knowledge. These concepts are similar to those proposed by Malone and Crowston (1994). The conceptualization of the e-business process universe of discourse for an intelligent infomediary-based e-marketplace is: In an e-business process, a business enterprise is represented by an agent to perform activities which are coordinated by resources. Description Logic Model for Knowledge Representation of E-Business Processes The elementary descriptions of the atomic concepts in the intelligent, infomediary-based e-Marketplace problem domain include: i. Business enterprise (BE) ii. Agent (Ag) iii. Business activity (Ac) iv. Resource (Rs) Elementary descriptions of the atomic rela- tionships in the intelligent, infomediary-based e-marketplace problem domain include: i. Represents ( { IsRepresentedBy - ) ii. Performs ( { IsPerformedBy - ) iii. Coordinates ( { HasCoordination - ) Here, if R is a relationship between two concepts in the problem domain, then R - denotes the inverse of the relationship R. DL derives its descriptive power from the ability to enhance the expres- siveness of the atomic descriptions by building complex descriptions of concepts using concept constructors. These terminological axioms make statements about how concepts or roles are related to each other. This develops a set of terminologies, comprised RI GH¿QLWLRQV ZKLFK DUH VSHFL¿F D[LRPV WKDW GH¿QHWKHLQFOXVLRQV) or the equivalence ({). The increased expressive power of the language is manifested in a range of additional construc- tors, including: R.C (full existential value restriction) ¬C (atomic negation of arbitrary concept) Figure 3. E-business process universe of discourse for an intelligent, infomediary-based, e-market- place Business Enterprise Agent Activity Resource Is Represented By Performs Represents Is Performed By Coordinates HasCoordination Owns a Business Enterprise Agent Activity Resource Is Represented By Performs Represents Is Performed By Coordinates HasCoordination Owns a 2442 Semantic Knowledge Transparency in E-Business Processes < n R (at-most cardinality restriction) > n R (at-least cardinality restriction) = n R (exact cardinality restriction) < n R.C (qualified at-most cardinality restriction)º > n R.C TXDOL¿HG DWOHDVW FDUGLQDOLW\ UHVWULF- tion) = n R.C TXDOL¿HGH[DFWFDUGLQDOLW\UHVWULFWLRQ < n R (concrete domain max restriction) > n R (concrete domain min restriction) = n R (concrete domain exact restriction) Given the aforementioned concepts and rela- tionships in the problem domain, we can begin to GH¿QHWKHUHODWLRQVKLSVEHWZHHQWKHFRQFHSWVLQ WKHGRPDLQ+HUHZHGH¿QHWKHWHUPLQRORJ\IRU the intelligent, infomediary-based e-marketplace problem domain using the following terminologi- cal axioms. This forms the knowledge represen- tation terminology, or TBOX, for the problem domain and the basis for the machine-interpretable representation of the ontology in OWL-DL. A BEFRQFHSWLVGH¿QHGDVDThing, the top concept in OWL-DL, which is represented by at least one Ag in the problem domain. BusinessEnterprise (>1 IsRepresentedBy Agent) (= 1 HasID StringData) (>1 HasAddress Address) (>1 HasDescription StringData) (>1 HasReputation StringData) (>1 HasTransactionSatisfactionHistory StringData) An AgFRQFHSWLVGH¿QHGDVD7KLQJWKDWUHS- resents a BE and performs activities for the BE. Agent (= 1 HasID StringData) ( = 1 Represents BusinessEnterprise) ( >1 Performs BusinessActivity) A Business ActivityGH¿QHGDVD7KLQJWKDW is performed by an Ag, has a coordination rela- tionship with Rs, and has a Begin Time and an End Time. Business Activity ( = 1 hasLabel StringData) ( >1 isPerformedBy Agent) ( >1 hasCharacteristics StringData) ( >1 HasDescription StringData) ( >1 HasCoordination Resource) ( = 1 hasBeginTime DateTimeData) ( = 1 hasEndTime DateTimeData) Each RsLVGH¿QHGDVD7KLQJWKDWLVRZQHG by exactly one BE and coordinates Business Activities. Resource ( = 1 hasID StringData) (>1 hasOwner Business Enterprise) (>1 Coordinates BusinessActivity) We utilize a novel and theoretically grounded, activity-resource, coordination mechanism for capturing the relationships between activities and resources. This allows for the explicit modeling of the coordination of individual business activi- ties, and the e-business process itself, using the information and knowledge resources in inter- organizational e-business processes over virtually integrated business enterprises. Business Activi- ties depend on resources and require coordination mechanisms to resolve these dependencies in an e-business process. A resource is related to an activity through a Coordinates relationship, where the resource coordinates business activities through various coordination mechanisms. Resource (Coordinates BusinessActivity) BusinessActivity (HasCoordination Re- source) 2443 Semantic Knowledge Transparency in E-Business Processes 'HVFULSWLRQORJLFDOORZVIRUWKHVSHFL¿FDWLRQRI generalization-specialization hierarchies of rela- tionships. We use the notion of activity-resource GHSHQGHQF\ZKHUHDFWLYLWLHVKDYHDVKDULQJÀRZ RU ¿W GHSHQGHQF\ ZLWK D UHVRXUFH 0DORQH HW al., 2003) to specify the relationships between activities and resources. Here we assume that the Coordinates relationship between resource and activity is an abstract, general relationship, which materializes in the form of the specialized relation- ships where a resource may coordination activities through a CoordinatesFlow, CoordinatesFit, or CoordinatesSharing relationship. Coordinates CoordinatesFlow CoordinatesFit CoordinatesSharing CoordinatesFlow CoordinatesFlowProducedBy CoordinatesFlowConsumedBy In addition, the CoordinatesFlow is further specialized to capture the activity-resource co- ordination where the resource coordinates the ÀRZRIDFWLYLW\E\HLWKHUEHLQJSURGXFHGE\RU consumed by a business activity. We utilize the previous inheritance hierarchy of the Coordinates relationship to develop a complex description of the relationship between Rs and Business Activities, as expressed in the following terminological axiom. Resource (>0 CoordinatesFlowProducedBy Busines- sActivity) (>0 CoordinatesFlowConsumedBy Busines- sActivity) (>0 CoordinatesFit BusinessActivity) (>0 CoordinatesSharing BusinessActivity) Information and knowledge are the primary resources pertinent to the problem domain we consider in this chapter. We utilize the concept GH¿QLWLRQV Resource Information Knowledge These complex descriptions of concepts, built from atomic descriptions, describe classes of objects in the problem domain and their inter-re- lationships. The terminological axioms presented previously make statements about how concepts and relationships are related to each other. The VHWRIWHUPLQRORJLFDO D[LRPV LQFOXGLQJ GH¿QL- tions, provide the terminology, or the TBox, for a SUREOHPGRPDLQ7KHDIRUHPHQWLRQHGGH¿QLWLRQV comprise the terminology for the intelligent, in- fomediary-based e-marketplace problem domain, LQFOXGLQJ WKH GH¿QHG DQG SULPLWLYH FRQFHSWV and the binary relationships between them. This provides the meta-level ontology and knowledge representation for the knowledge base in an intel- ligent, infomediary-based e-marketplace. The other component of the knowledge base, in addition to the terminology or TBox, is the world description, or ABox that includes descriptions of individuals in the problem domain. Together, the TBox and the ABox comprise the knowledge representation system based on description logics. The knowledge representation system provides the knowledge base and facilities to reason about the content. Proposed Semantic Knowledge Representation for Supplier Selection for Infomediary-Based E-Marketplace An e-procurement, supplier selection, e-business process in an infomediary-based e-marketplace requires information of attributes that describe the buyer’s requirements, such as price, quantity, and the date by which the item is required. The selec- tion of a supplier, from a set of suitable suppliers . information and matching buyers’ needs with sellers’ products and/ or services. Grover and Teng focus on the critical information-providing role of the market and identify the roles played by electronic. Malone, Crowston, and Herman (2003) and are similar to those in Van der Aalst and Kumar (2003). The complexity of coordinating e-business processes and the increasing demand by customers for. activities and their relationships, cri- teria to indicate the start and the termination of the process, and information about the individual activities, including participants and data (www. wfmc.org).