Seite - 173 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

call ofoperationsbymessagepassing,whichmeans thatonly this specificaspect canbe measured using ametric scale.However, this aspect can be embedded into the existing classification approaches, so that all other aspects of interoperability are still hierarchi- callymeasurable, but this one specific aspect offers a complementarymetricmeasure- mentmethod.Toembedtheaspectwhichiscoveredbyourmetricmeasurementmethod, weuseaclassification frameworkwhichhasbeendesignedbyPersisterasandTarabanis in addition to the twelve approacheswhich they have analyzed.TheConnection,Com- munication, Consolidation, and Collaboration Interoperability Framework (C4IF) uses anevolutionaryperspectiveaswell. It defines the following levelsof interoperability: • Connection: Includes theaspectsofbeingable toexchangesignalsbetween two informationsystemcomponents. • Communication:Theabilityofcomponents toexchangeanykindofdatawhich is conform to a predefined data format agreed by both components. This level encompasses twoaspect sub levels covering thedata formatof single exchanged strings ([6]mentions adata format for a stringwithdd/mm/yyyyas anexample) on thefirst level aswell as the data format ofwhole structures (e.g. an JSONor XMLschema)on the second level. • Consolidation: Includes the aspect of being able to understand exchangeddata. This requires that both components share and agree on the same meta model describing the semanticmeaningofa single stringandwholedata structures. • Collaboration:Thehighest levelof interoperability includes theabilitynotonly to understand themeaning of exchanged data but also to use the data in such a way that whole processes andworkflows can be formed by triggeringmultiple functions and services. [6] describes this level as “the ability of systems to act together”.Theseactions should triggerchanges in thephysicalworld. According to Persisteras and Tarabanis, the C4IF has been developed by transfer- ring somewell-established concepts from linguistics into the concepts of information system interoperability.More in detail, these four levels of interoperability aremapped to equivalent levels of the human communication. Persisteras andTarabanis argue, that these concepts of linguistics are considered as common to any kind of communication scenarioandcan thereforebeapplied toallkindof interoperability scenario in the infor- mationsystemcontext [6].For the fulfillmentofour interoperabilityaspectof issuingan operation by sending amessage,we assume that the aspects of theConnection aswell as of theCommunication levelmust be fulfilled so that an operation can be issued by sendingamessage.Nevertheless, it is required that the receiverof themessage isable to understand itsmeaning inorder to choose andperform the appropriate operation. Since the terminology “message passing” can be considered as equal to the terminology of “exchangingdata”, theaspect isplaced in theConsolidation level. 4. Metric InteroperabilityMeasurementModel Our approach tomeasure the degree of interoperability as ametric quantity is inspired by the observation of communicating processes within the consolidation level of the C4IF classification scheme. In the following,wewill decompose this behavior in order toextract relevant componentsof themodel. S. Kotstein and C. Decker /AnApproach for Measuring IoT Interoperability 173