Career

A Multidisciplinary Framework for Concept Evolution: A Research Tool for Developing Business Models

Description
The paper describes a new framework for multidisciplinary concept evolution (MCE). The impetus for systematizing the concept was a practical need to facilitate successful communication between different disciplines. The main benefit of the framework
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
   Abstract — The paper describes a new framework formultidisciplinary concept evolution (MCE). The impetus forsystematizing the research concepts was a practical need tofacilitate successful communication between differentdisciplines. The main benefit of the framework is that it aidsthe complex and dynamic process of conceptualization byhighlighting abstraction, generalization, and ontologyengineering as practical methods to implement conceptevolution. One application of the framework is in solvingcomplex business-related problems. The MCE framework canbe utilized not only by researchers but also by othercommunity stakeholders. Applying the framework to scientificdisciplines may bring additional value to research as well asbenefits to practical development endeavors.  Keywords — Business models, multidisciplinary research,concept evolution, semantic interoperability. I.INTRODUCTION  A.General The broad success of the Internet has led to theemergence of eBusiness (eB). Although eB has opened newpossibilities for enterprises (especially for SMEs), itsspreading popularity has partially influenced the tighteningof competition in the global market environment. Facinglimited resources, companies have tried to achieveeconomies of scale by establishing closer relationships withother companies in the form of business or enterprisenetworks. By networking, companies wish to save costs andachieve synergy effects in business-to-business (B2B)operations. This can be achieved, for instance, by usingintegrated information systems. These systems, whichenable the distribution of network resources, are based onprocess thinking. This process view calls for a new“business thinking”, which utilizes cross-organizationalchains of business processes. This increases demands formutual understanding and trust, which are prerequisites forimplementing these cross-organizational businessoperations. Following jointly agreed procedures andcontracts is one way of enforcing mutual and multilateralco-operation. However, the difficulties in combiningdifferent views and opinions still remain as a considerableobstacle in finding an agreement – even within a singlecompany.  B.Background: Project Description The demands of the modern business environmentdescribed above form the basis of work carried out in aresearch project focusing on integrating enterprisenetworks. The project, which embraces the interrelatedresearch areas of business models, systems integration, andinformation security, seeks to analyse and modelinformation, material and financial flows in the enterprisenetwork, in an effort to combine all the individual results ina general model of integrated enterprise network.Although they are all part of the common researchproblem, the focal points of these research areas are slightlydivergent, yet complementary. Business models researchaims to develop a new business network model whichintegrates processes in small and medium-sized enterprises(SMEs). The need to integrate the business processes hasincreased with the digitalization of B2B operations. Thegoal of systems integration research is to define anelectronic business process collaboration prototype model.The developed business process integration model can thenbe used as a basis for service concepts and for a class of architectures offering a solution to B2B integration.Information security in this context is needed to implementbusiness process integration safely. C.Research Setting The development of business-related models is in thescope of design sciences, as stated in March and Smith(1995). Design science attempts to create things that servehuman purposes, whereas natural science tries to understandobservable reality. Business models clearly serve humanpurposes. March and Smith (1995) continue with the fourtypes of design science products: constructs, models,methods and implementations. They argue that as in naturalscience, there is a need for a basic language of concepts (i.e.constructs). These concepts make it possible to characterize A Multidisciplinary Framework for ConceptEvolution: A Research Tool for DevelopingBusiness Models Jukka Aaltonen 1 ; Jukka Rinne 2 ; Ilkka Tuikkala 3   1    Researcher, University of Lapland, jukka.aaltonen@ulapland.fi 2  Researcher, University of Lapland, jukka.rinne@ulapland.fi 3    Researcher, University of Lapland, ilkka.tuikkala@ulapland.fi  phenomena. March and Smith state that these constructscould be combined into higher level constructions (models)which are used to describe tasks, situations, or artefacts.In accordance with the design science view, constructiveresearch (CR) was chosen as the main researchmethodology, largely for practical reasons. CR is generallyseen as a case-study method which aims to find solutions(constructions) to predetermined problems. If viewed as anapplied research method, the essential feature of CR is thegeneration of new knowledge of the target area. Accordingto Kasanen, Siitonen and Lukka (1993), finding apractically relevant research problem, obtaining a generaland comprehensive understanding of the topic, andinnovating and constructing a theoretically groundedsolution are crucial steps in the constructive researchapproach (CRA). Although a relevant research problem canbe discovered from a purely theoretical basis, it is morecommon to find actual research issues from premises of existing real-world business demands (Labro and Tuomela,2003). In the final stages of the CRA, the developedconstruction should be evaluated and tested through anexamination of its applicability and an illustration of itstheoretical connections and research contribution (Kasanenet al., 1993).Following the CRA stages from the standpoint of designscience and surveying the existing theories and concepts, itbecame evident that there are severe distinctions in theconceptualizations between each research area. It was seena necessity among researchers to find a ground-levelconsensus of at least the essential concepts like: businessprocess, business modelling, electronic commerce, B2Bintegration, trust etc. Additionally, the complexities inselecting the concepts in the scope of the research and thechallenges in communicating the definitions thereof between the different parts of the research, acted as adriving force for developing a formal framework formultidisciplinary conceptualization.  D.Structure of the Paper  The next section explains how the challenges introducedby complex cross-disciplinary concept semantics led to theidea to develop a generic framework for multidisciplinaryconcept evolution (MCE). After this, the practical methodsof concept evolution and the notion of emergence of meaning are briefly illustrated. Building on these findings, aproposal for how the framework can be utilized in solvingcomplex business-related problems is then described,followed by a concluding discussion about the usefulness of the developed model.II.CHALLENGES OF MULTIDISCIPLINARYCONCEPTUALIZATIONThe above issue of concept diversity experienced in theproject can be illustrated, for example, by usingterminology adapted from the well-known methodology of formal concept analysis (Wille, 1982), which uses amathematical notion of lattices to represent the relationbetween concepts (or objects) and their properties (orattributes). The philosophical background of formal conceptanalysis (FCA) lies in the definition of  concept  (of a givencontext) as a unit of thoughts consisting of two parts, the extension and the intension . The extension covers allobjects belonging to this concept and the intensioncomprises all attributes valid for all those objects (Wagner,1973). In relation to business network research, the initialassumption of a shared concept context can now beformulated using the practical guideline for FCA (Wolff,1993): the extent of the topmost concept, i.e. the mostgeneral super-concept of the combined research domain,should always be the set of all concepts; its intent does notcontain any properties (in the mentioned context of theproject’s research domain). However, the added complexityof the domain of the discourse, caused mainly by theprojects multidisciplinary approach and the diverging viewsin each research area about details of the concepts of eachdomain, required a more sophisticated treatment of conceptsemantics.Additionally, the overlapping nature of the conceptdescriptions is exemplified in how the three research areashad in many cases inconsistent concept definitions: on theone hand, there were concepts that had a uniform (or nearlyidentical) name but they still referred to a completelydifferent (real-world) entity, or they had a contradictorymeaning between research views; and on the other hand,some concepts with unrelated names appeared to beidentical in their definitions or referred to the sameunderlying entity.In the context of the initial project’s collective researcharea, these findings about the complex interplay of closelyinterconnected and at the same time divergent views of three overlapping domain concepts (business models,information security and systems integration) prompted usto discover an intrinsic layered structure of conceptcomposition, which is illustrated in Figure 1. Threeoverlapping areas of concepts can thus be distinguished: (S)a single-topic layer; the concepts appearing in only oneresearch domain, (D) a dual-composite layer; concepts thatfall within the intersection of two research domains, and(M) a multi-composite layer; a set of concepts that fallwithin the intersection of all research domains.By means of generalization, an extension of the describedmodel is also depicted on the right side of the diagram.Here it can be seen that the overlapping nature of cross-disciplinary conceptualizations presents itself as a multi-layered structure, where each layer contains a certain set of domain concepts (srcinating from one or several domains)according to their interconnectedness (that could presentitself as a similarity or as relevance, but also as dissimilarityor even as being conflicting) with the other domain’sperception of the entity represented. Using this kind of   semi-analytical abstract modelling in each research areaindividually (or in collaboration), the domain concepts canbe categorized and related not only according to their“internal” properties but also along their “external”relations to other research views. It can be claimed that this way of thinking differs fromthe more traditional conventions used in multidisciplinaryresearch, which have shown a tendency to only try toconnect, combine or converge the already existing conceptmodels of different disciplines in an attempt to build acoherent representation of the whole.An additional discovery was made when it was realizedthat formalizing and further developing this kind of arepresentation could serve as a metamodel. As such, thedeveloped framework can be seen as an explicit model of constructs and rules needed to build specific models withina domain of interest. Thus, the described conceptualizationschema could further be developed in accordance with thepractices of metamodelling as, for example, in thespecifications under the Model Driven Architecture (MDA)by the Object Management Group (OMG; Model DrivenArchitecture). The following sections focus on advancingthe construct in the direction of metamodelling only to theextent that metamodelling can be used as an informaldescription that is robust enough to present the essentialbuilding blocks of the proposed practical researchframework.Even in its present informal state, this cross-domainconceptualization schema could already be used in theinitial research project in combining and harmonizing thewhole concept domain in such a way that the differentconcept definitions and descriptions are also conserved.III.CONCEPT EVOLUTION AND THE EMERGENCEOF MEANINGThis section describes the essential constructs andpractices that need to be incorporated into the previouslydescribed preliminary metamodel for it to be useful inactual research. The main goal here is to present thebuilding blocks of a framework for multidisciplinaryconcept evolution (MCE). Figure 1: Layered composition of domain concepts  The fundamental functional requirement of a practicaldesign tool in this context is that researchers should be ableto use it to generate a coherent, formally expressive andmutually accepted representation of strongly inter-connected research domain concept semantics. This kind of compilation of knowledge is here called the general and shared body of meaning . In accordance with the dictionarydefinition (Oxford English Dictionary), the emphasis here ison the emergence of an understanding about thesignificance, purpose or the underlying truth of the observedand analyzed conceptualizations. The justification forintroducing meaning to this discussion lies in the insightthat it is a valuable notion in tackling autonomy andheterogeneity issues (for example, in the dynamicintegration of information systems) and to enable solutionsto general problems in social, pragmatic, semantic andsyntactic interoperability (Ouksel, 1999, Open SystemsFramework for Social Interaction). Also the importantcommunity view is included in the framework byrecognizing that the quality of the generated repository of meaning is ultimately evaluated by its ability to convey thenecessary knowledge in an appropriate form to resolveconflicts, uncertainties and misunderstandings between the stakeholders or agents participating in the real-worldphenomena being studied.It is proposed here that mostly the operationalrequirements can be fulfilled by applying (i) abstraction ,(ii) generalization and (iii) ontology engineering (particularly ontology mapping) to enable the emergence of meaning. However, before these methods or operations aredetailed, the novel idea of  concept evolution is defined asthe concrete activities performed by the researchers thatgive rise to a higher-level representation emerging from theinitially constructed conceptualization.As illustrated in Figure 2, all the mentioned metamodelcomponents, the overall description of the listed practicalmethods and the basic ideas about semantic interoperability,in combination with the given functional requirements,together constitute the main building blocks of the MCEframework.As stated before, the MCE framework identifies thefollowing practical methods to aid in the complex anddynamic process of concept evolution:i. generalization : used here in the same sense as in thewell-known object-modelling and set-theoryparadigms: inheritance (is-a relation),subsume/supersume, object attribute relations andassociations; for example, Formal Concept Analysis(FCA). Figure 2: Concept evolution and emerging meaning  ii. abstraction : seen here as a creative cognitiveprocess, during which subjective interpretations aremade (with the participation of domain specialistsand experts, in addition to researchers) about thephenomena being researched and the initial conceptsused. Abstraction (as the common-sense meaningsuggests) can help the emergence of novel conceptsby reducing the level of detail and by enablinghigher-level representations.iii. ontology engineering : based on the definition of  ontology as appropriate for the Semantic Webinitiave and as specified in the OWL specification bythe World Wide Web Consortium (W3C; WebOntology Language). Ontology mapping (Kalfoglouand Schorlemmer, 2005) is an especially usefulpractice here mainly in providing formal descriptionsand tools to glue research-originatedconceptualizations together. Additionally, it can alsobe used to relate or link evolving concepts andmeanings to various external knowledge repositories .These repositories lie outside the actual researchdomain proper and may include things like: relatedexisting ontologies, knowledge repositories,vocabularies, dictionary definitions, taxonomies,standards and enabling technology recommendations,generally accepted naming conventions, code lists,etc.More formally these operations (i.e. the practical meansof concept evolution) can now be expressed as a mapping(or a function), the domain of which consists of the sets (S i ,D i,j , .., M i,j,k,.. ) of overlapping domain area conceptualizations(not the individual concepts), and whose range is theconstituents and the representations of meaning (m i ).Summing up the benefits of using the framework,researchers are able to identify and analyse concepts fromall relevant research areas and to categorize them intodifferent classes. The framework also makes it possible togenerate hierarchical concept schemas and to classifyvarying concept descriptions and definitions. Additionally,from the perspective of information system design, theframework is useful in providing at least a semi-formal (i.e.possibly machine processable) model of the domain of interest. It must be noted, however, that the practical use of the MCE framework in a variety of research situations stillrequires, for one thing, the specification of a formal methodof utilizing the basic ideas and constructs discussed here.IV.MCE FRAMEWORK IN THE CONTEXT OF NEWBUSINESS MODELSWhen applying the MCE framework to a businesscontext, an examined business-related phenomenon forms astarting point for utilization. When it is examined from theviewpoints of different business stakeholders andresearchers, the outcome is a collection of different conceptdomains, which paradoxically are not compatible with eachother. This diversity of concept domains has traditionallyled to deviating views on research issues, because the targetphenomena is interpreted and described in subjectiveconventions. The MCE framework makes it possible toachieve general and shared meanings for concepts of focused phenomenon. These meanings can be adopted byall the research areas in question, and can further be used toform subjective constructs that share their conceptualdefinitions. This is a crucial aspect in designing orformulating new business models. Business phenomena aretypically complex, involving features from differentdisciplinary backgrounds.In this research project the specific business-drivenphenomenon is an integration of business processes, whichis expected to lead to a new business model (integratedenterprise network). As Figure 3 illustrates, by processingthe initial concepts “business process”, “integration” and“business network” in the MCE framework, a common setof meanings is attained. This collection is then used in thenext (construction building) phase of the CRA as a valuableresource to aid in the creation of new solutions (constructs)for the described research problem. Developed constructscan further be combined into new business models. Forexample, a secure integration of business processes in anintegrated business network demands a combination of constructs from different research areas. A prerequisite forthis is that there exits general, shared meanings for theconcepts of all related research areas.One of the important consequences of applying the MCEframework to business process integration research issues isthe insight that the study of information security should atleast now be accepted as an important area of modernbusiness. Because of the fact that information security is abusiness issue, not a technical issue (adapted: von Solms,1999), a failure in designing and implementing it hasevident effects on the ability of enterprises and networks tofunction. A severe information security incident can lead togrowing insecurity and erosion of trust, which dissolves anetwork’s capability to function. Information security ismore than a technical issue: it is also an organizationalissue. Additionally, systems integration research should alsobe added to the list of business issues (Anderson, Longley,Kwok, 1994), and should thus be investigated in parallel toother research areas. A modern business relies heavily oninformation systems, which can be interconnected by themeans of systems integration. IT is also one of the keyenabling forces in networking development, because itenables a cost-effective way for enterprise interoperability.In contrast, as a standalone research approach, systemsintegration mainly provides the functional service view asan infrastructure for implementing business operations atthe organizational level.These distinct conceptualizations inflict difficulties incommunication between different stakeholders. However,
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks