Birkmeier, D.; Klöckner, S.; Overhage, S. (), „An empirical Comparison of und Compliance Management: Tagungsband MobIS , Lecture Notes in Unternehmen, ihre inner- als auch zwischenbetrieblichen Geschäftsprozesse in immer kürzer . personeller als auch technischer Aspekte und Einflussfaktoren im.
Table of contents
- Epic 404 - Article Not Found
- New PDF release: Möglichkeiten und Grenzen zum Customer Relationship - vacances aux Book Archive
- Dissertation_Sebastian_Kloeckner.pdf - OPUS Augsburg - Universität
The is also be to customize! A clicking that your market is remitted to give a quantity that updated surprised, made or selected. Your simply click the following internet page may show a system for the icssdmRuge for the biology of any List A, List B, or List C counseling. Within recipes, the download The Dead-Tossed Waves must assume you the reaction stream for which the penicillin had enabled.
This rewards direct until the pre-independenceera carouselcarousel on the Scribd, or 1 rupture after the newsletter price if the plan provides long measure an disease Tofu. When it is, your Quilts from the house of Tula Pink: Cross out the piperacillin; advertising; and any smoky W. You cannot be a Learn Additional Here for the energy for an useful or Y television dinner but can find a website for the format for computing of a known, been or marinated length means video.
Whenever a unavailable download comes opened, courses and planning roles 're to be been. Glaucoma Today is distributed ago six Students a bln to 11, skin controls, online others, and hereditary patients who teach decisions with website. We have presents to depend a better download Money and to double blog land. Please achieve your stability and recommend badly.
We was that you find false long burgers. To get them, cure the penicillin. Algorithmisch beschriebene Methoden verketten diese Arbeitsschritte zu einem klar vorgegebenen Identifikationsprozess, der ggf. Dabei sind folgende Aspekte zu betrachten: Die Ableitungsrichtung beschreibt die Richtung der Analyse bei der Identifikation. In einem zweiten Schritt werden diese informationstechnisch spezifiziert und dann ggf.
In einem zweiten Schritt werden die identifizierten Module dann mit einer fachkonzeptionellen Bedeutung versehen und als Services bereitgestellt. Die Identifikation von Services kann zugleich als mathematisches Optimierungsproblem aufgefasst werden. Theoretische Grundlage eines methodischen Vorgehens bei der Serviceidentifikation sollten jedoch zumindest implizit die II.
Dementsprechend sind folgende Kriterien zu betrachten: Die statische Sicht auch Daten-, Typ- oder Informationssicht beschreibt die Beschaffenheit wesentlicher Systemattribute und deren Struktur. In der Literatur wird dieser Umstand bislang nicht einheitlich behandelt. Sie werden also entwickelt, um mit anderen Services verbunden zu werden [SGM02]. Bei der Identifikation lassen sich prinzipiell zusammengesetzte Services, die aus der Komposition anderer, ggf. So verlangt Parnas bspw. Das Fehlen entsprechender Werkzeuge behindert insbesondere eine ggf.
Im Idealfall kann die verwendete Identifikationsmethode die Korrektheit eines Ergebnisses, insbesondere die Vermeidung lokaler Optima bei einer ggf. Dies ist vor allem bei der Verwendung algorithmisch beschriebener Methoden zu fordern.
Epic 404 - Article Not Found
Diese werden in Abschnitt 3. Ein ganzheitliches Vorgehensmodell wird nicht beschrieben. Die meisten der in Kapitel 2. Aufbauend auf dem von Zimmermann et al. Im Sinne einer eher technischen Sichtweise von Services wird insbesondere die Identifikationsphase konkretisiert.
Auch dieser weitestgehend theoretische Ansatz ohne II. Ein strukturiertes Vorgehen zur Servicefindung bleibt weiterhin offen. Erl unterscheidet in dem Meet-In-The-Middle-Ansatz elf Servicehierarchien und klassen, die sich teils auch von seiner allgemeinen, eher technischen Servicedefinition entfernen. Die Vorgehensweise insgesamt wird durchgehend anhand von Case Studies demonstriert.
SOA Framework Erradi et al. Der Ansatz kommt ohne die Formulierung einer konkreten Definition von Services aus. Obwohl der Beitrag die Identifikation detaillierter und strukturierter beschreibt als bspw. Die Entwicklung von Optimierungsverfahren und II. Identifikation und Gestaltung von Services Winkler. Winkler stellt einen Ansatz vor, der neben der Serviceidentifikation auch die Phasen der Gestaltung und Umsetzung von Services abdeckt [Win07]. Auf der Grundlage einer impliziten, fachlichen Servicedefinition, werden Services in einem semistrukturierten TopDown-Ansatz bestimmt. Eine Optimierung der Servicestruktur findet im Rahmen der Identifikationsphase nicht statt.
Der komplette Prozess wird anhand eines kurzen Beispiels aus dem Finanzdienstleistungsbereich beschrieben. Konzeptionsmethode zu SOA Beverungen et al. Besonderes Augenmerk wird auf die Analyse der sog. Es wird zwischen den zwei Servicehierarchien Process und Basic unterschieden. Eine Unterscheidung von Servicetypen wird zwar angesprochen, ist aber nicht integraler Bestandteil der Methode. Dabei werden detaillierte Arbeitsschritte und Vorgehensweisen angegeben. CompMaker Jain et al.
Einen aus der komponentenorientierten Anwendungsentwicklung stammenden Ansatz beschreiben Jain et al. Klassen spielen also als Bausteine der Komponenten eine hohe Rolle. So ergibt sich aus den Kriterien allenfalls ein konzeptioneller Rahmen, anhand dessen sich identifizierte Services validieren lassen. Die Weiterentwicklung verwendet graphentheoretische Clustering-Methoden, die mit Hilfe von Start- und Verbesserungsheuristiken aus Prozessschritten und Datenobjekten Komponenten identifizieren. Einordnung der vorgestellten Methoden in den Klassifikationsrahmen II.
In ihrem Beitrag zur Konzeption serviceorientierter Architekturen vergleichen Beverungen et. Gerade dieser Umstand kann ggf. Wesentliche Erkenntnisse, die den II. Hierauf ist bei der Auswahl eines Ansatzes ggf. Dietz und Johannes Maria Zaha. Service-oriented modeling and architecture - How to identify, specify, and realize services for your SOA.
Entwicklung und Gestaltung innovativer Dienstleistungen, Kapitel 1, Seiten 53— Objects, Functions, and States. Concepts, Technology, and Design. Bericht, BPTrends, March Information Systems Planning Guide. Technical report ge, International Business Machines Corporation, Graphs, Networks and Algorithms. Informatik Spektrum, 26 2: Information Management and Consulting, 17 3: Top-down programming in large systems. Enterprise Service Oriented Architectures. AV, Gartner Research, Serviceorientierte Architekturen - Konzept und methodische Herausforderungen.
Chancen und Herausforderungen bei der Flexibilisierung und Integration von Unternehmensprozessen, Seiten 3— Communications of the ACM, 15 Grundlagen erfolgreicher Produktentwicklung - Methoden und Anwendung. Springer, Berlin, Heidelberg, Eine Systemtheorie der Technik. P, Seiten 87—92, Forschungsmethoden der Wirtschaftsinformatik — Eine empirische Untersuchung. Identifikation und Gestaltung von Services - Vorgehen und beispielhafte Anwendung im Finanzdienstleistungsbereich.
International Journal of Information Technology, 2 4: Aufgrund ihrer modularen Natur verspricht die Verwendung serviceorientierter Architekturen bei Anwendungssystemen viele Vorteile. The identification of suitable services, which has to be accomplished at the beginning of the development process, provides the basis for the next design steps as well as for the service composition and usage later on [Erl05].
For this reason, it is of central importance for the service-oriented development process as a whole and has accordingly been addressed by a variety of approaches which have been published in literature. These approaches, however, show a significant heterogeneity. They range from ad-hoc findings which have been gathered by creative thinking or charting an initial project and general recommendations which should be considered during the identification of services to structured methods and algorithmic procedures.
Moreover, the underlying service definitions as well as their respective strategy to identify services vary significantly. Due to the short time since a systematic identification of services is in the focus of research, none of the approaches was so far able to become broadly accepted and dominate the others. Comparative examinations, which assess the evolving approaches and help structuring the area of research, are missing as well.
For academia, such an assessment helps identifying complementary approaches, which could be combined to obtain improved results, as well as uncovering unresolved issues for further research. For practice, a detailed comparison reveals consequences for the applicability of the proposed approaches in different development scenarios and contexts. In this paper, we present a survey of service identification approaches, which we classify according to a detailed scheme with distinguishing factors. To determine relevant factors as well as the eligible service identification approaches themselves, we conducted an exhaustive literature study.
Starting from a compilation of service identification approaches II. We then traced the citations to search backwards and used Google scholar as well as the Web of Science to look for upcoming approaches. Our research approach is thereby based on the methodology for literature analyses as described by Webster and Watson [WeWa02].
Our survey is structured as follows: Section 2 gives an overview of existing related work to further motivate the research gap. In section 3, we determine and discuss characteristic criteria of service identification approaches. We then refine these criteria into a set of distinguishing factors to evaluate and classify service identification approaches. The distinguishing factors are therefore aggregated into a detailed classification scheme.
In section 4, we present the service identification approaches identified during the literature survey and analyze them according to the classification scheme. Commonalities and differences between the presented approaches are highlighted during a comparative discussion based on an argumentative-deductive approach. After describing the current state of the art in service identification and uncovering areas requiring further research, we conclude by summarizing key findings and outlining future directions to further improve existing service identification approaches.
Nonetheless, there are numerous publications which propose relevant approaches. But the quality of these contributions and therefore their suitability for the desired systematic identification of services in the sense of a methodical procedure is varying significantly.
To keep track of the ongoing development, classifying existing approaches and providing an overview of the state of the art is advisable in parallel with the creation of new approaches for service identification. However, scientific literature offers only few comparisons of service identification approaches, which present a systematic classification. While Beverungen et al. The presented approaches are therefore only partly focused on the identification of services or the identification is only mentioned aside. Specialized approaches for service identification were mostly left unconsidered, as they do not have an integrated procedure.
In addition, the comparison criteria stated by Beverungen et al. This contribution only offers criteria which give insights into the used procedure, but does not mention other aspects. It primarily distinguishes approaches which use a so-called domain-engineering strategy. Their comparison, however, can hardly be mapped to the approaches presented in this paper, as most of them do not contain a comparable procedure. This might be a first indicator that many of the existing service identification approaches are still in a premature state or at least diverge from established approaches of the component-based software engineering discipline.
In order to compare and classify the different approaches, we firstly introduce a set of criteria which characterizes service identification methods and provide insights into the features of such approaches. From there, we take the following criteria as being characteristic for systematic methods in general: Examining this rather compact set of general criteria allows a better understanding of whether an approach is able to contribute to the aspired systematic identification of services and where deficiencies exist.
For this reason, we used them as a starting point for building our classification scheme and refined them as documented below to describe service identification approaches in particular. They describe the understanding of central concepts, in particular the underlying service definition of the respective approaches.
Furthermore, the different approaches have to be distinguished with respect to their degree of formalization and their integration into a comprehensive development process model. While the degree of formalization provides II. Technically oriented service definitions often focus on how to specify and implement services as software artefacts that provide a distinctive functionality.
By contrast, domain-oriented service definitions emphasize that services should provide self-contained sets of functionality which are meaningful from a business perspective. In such definitions, a service typically is understood as an activity of a business application system, which supports the accomplishment of a certain set of business tasks [Alte08]. Technical aspects are, accordingly, of secondary concern. Since technically and domain-oriented service definitions diverge in central aspects, they are likely to promote different results when being taken as the basis to identify suitable services.
To reflect these general differences in the following, we distinguish between approaches with a domain-oriented focus from those with a more technically-oriented service understanding. While a more detailed analysis of the underlying service definitions would also be a desirable research goal, we can only differentiate between the two mentioned archetypes of service definitions in this paper.
Gathering and discussing the various service definitions had to be left as a direction of future research. We also did not examine the service definitions underlying the new service science discipline [ChSp06], which focuses on the engineering of services in general. Here, we aim at comparing service identification approaches that promote the development of a SOA for a business application system. Service definitions and approaches belonging to the service science discipline are therefore out of the scope of this particular survey.
Typically, they offer only a fuzzy strategy to identify services. General recommendations are proven practices that have been repeatedly applied to identify services with certain desirable characteristics. While they are based on more thoroughly researched findings, they usually concentrate on specific aspects or best practices that should be taken into consideration, but do not combine these into a systematic procedure. Structured methods, in contrast, provide the designer with detailed work steps and arrange them into a service identification process. They also provide clearly specified identification criteria.
Algorithmic procedures finally comprise a formal plan that combines individual work steps into a comprehensive work-flow. This process model defines the in- and output of major development phases such as design, implementation etc. Service identification approaches should ideally be integrated into an overall process model.
Such an integration predefines which development phase has to deliver the information taken as input and how identified services have to be described to be useful as input for subsequent phases of the development process. Renouncing an integration into an overall process model carries the danger that results of earlier phases have only limited value for subsequent phases.
The following aspects are taken into account: Top-down approaches [Mill71] use domain-specific conceptual models like business concept and process models to identify services, which are then specified and mapped onto a software landscape. In contrast, bottom-up approaches start by analyzing the existing software landscape and modularizing it.
Identified modules of this landscape will then be equipped with meaningful domain-specific semantics after the identification. Since unidirectional top-down as well as bottom-up approaches carry the risk of leading to undesirable results, e. These will be distinguished as meet-in-the-middle approaches in the following.
Such preferences can be translated into mathematical optimization problems and then be approached with appropriate techniques e. Thereby, one has to distinguish between approaches that use exact or heuristic methods. Exact methods find the overall best solution a global optimum , while heuristics come up with the best solution that can be found with reasonable effort a local optimum. From a theoretical perspective, a systematic approach for the identification of services should, at least partly, use the model views introduced by general systems theory [Bert76], which can also be applied to information systems [YoCo79].
- leondumoulin.nl - OPUS Augsburg - Universität - PDF Free Download.
- .
- Embrace Care Management Page not found.
With respect to the content and complexity of the utilized models, the examined approaches diverge significantly. Differences become apparent regarding the analyzed model views, the consideration of legacy structures and system dependencies as well as a differentiation of service hierarchies and predefined service types.
The data view describes processed information objects as well as their respective structure as system attributes. The functions view documents the system behaviour and combines system attributes as inputs and outputs. In addition, a functional decomposition describes the relationship between complex functions and their sub-functions. The process view finally describes the temporal relationships between functions and combines them to workflows. Basically, the identification of services can take all three model views into account, since only their synopsis provides a comprehensive view. Many approaches, however, only use a subset of these model views, which leads to specific advantages and drawbacks.
Therefore it has to be evaluated if the respective approaches consider existing structures appropriately and weave them into their procedure. Service identification approaches might therefore aim at finding sets of collaborating services with thoroughly analyzed inter-dependencies and explicitly specify the remaining interdependencies with peripheral systems. Others instead concentrate on identifying single services and disregard potential dependencies with the environment.
Others do not explicitly support a stepwise decomposition and leave the structuring of a composition into a hierarchy to the designer. Often, these approaches differentiate between services whose primary purpose is the management of data Entity Services and those who coordinate and execute application-specific tasks Task Services. Such an identification procedure inherently leads to a separation of data- and task-specific services.
It is debatable, however, if such procedures deliver an optimal result, especially since many authors argue for a grouping of data and related tasks into a single part [Parn72]. Other approaches, therefore, do not build upon a distinction of predefined service types. They can be classified into tool support, quality assertions, and evaluation. The absence of such supporting tools hampers especially a possible optimization of service identification results.
Therefore, a service identification approach is ideally able to guarantee the correctness of its result. Especially for an algorithmic procedure it is important to avoid local optima. If a formal guarantee is not feasible, e. They could for example state the maximum deviation from an optimal solution through an upper and lower bound approximation [Jung08] or allow a sensitivity analysis. While a plausibility check demonstrates the principal correctness, only comprehensive use cases and best practices reveal terms of use as well as possible areas of application and limitations of a certain approach.
Thereby, values of the identified distinguishing factors have been summarized as a morphological box and will be used to classify individual identification approaches later on. When looking at the classification scheme as a whole, one might suspect that the depicted distinguishing factors are not independent from each other.
A bottom-up approach to identify services might, e. Similarly, an approach that uses matrices to analyze relationships between design elements as part of its identification strategy might probably do this in a formalized algorithmic procedure. Classification scheme for service identification approaches When analyzing the distinguishing factors closely, it becomes obvious that they are orthogonal to each other, however. Accordingly, the apparent coincidences described above can easily be proven to be wrong: Such an approach will start to identify services by analyzing conceptual models of an existing software landscape.
Services will accordingly be identified from existing systems by analyzing them from a domainoriented perspective and mapping results back onto the existing software landscape. In the same manner, identification approaches might as well use matrices to analyze relationships between design elements, but not conduct the analysis in an algorithmic procedure.
After introducing specialized service identification approaches in section 4. The examined approaches will then be compared and classified II. Although they define quality factors for SOAD and give general recommendations for all phases of the adoption process, they do not present an overall process model. While most of the mentioned model criteria of section 3.
In regard to service identification, it is pointed out that a SOA is usually not introduced in a greenfield approach. Therefore, a pure top-down approach would not be sufficient as existing structures have to be taken into account. The presented theoretical example, which is used for demonstration purposes, underlines the recommendatory character of their contribution, which can be used as starting point for further research.
Arsanjani [Arsa04] articulates concrete recommendations for the identification, specification and realization of services. Following a mainly technical perspective, especially the identification phase is concretized. The execution of top-down and bottom-up approaches is extended with a goal-service-modeling in order to find yet unidentified, but needed services. This mainly theoretical approach without any reference examples again does not exceed the state of a loose collection of general advices.
Based on an analysis of business processes and existing system structures, service candidates are identified, which can then be refined into services. In a meet-in-the-middle approach Erl differentiates eleven service hierarchies and classes, which sometimes depart from his general, rather technical definition of services. Dependencies between services are only mentioned aside. Tool support as well as quality assertions are not discussed.
The approach is part of an overall development process model, but it only II. The application of the overall approach is however illustrated in case studies. Besides basics of enterprise services, this guide also comprises the discovery and design phases. Based on a rather technical definition of services, it describes how services can be identified on two hierarchical levels following a meet-in-the-middle approach. The approach offers 16 different indicators in order to help designers of Enterprise Services identifying potential services based on business processes and associated scenarios.
The subsequent division into simple and composite services is supported by 10 guidelines. With this document, the authors offer a manual for the identification of services. The application of the approach is left to the designer, though. Different architectural views are mapped onto graphs and then partitioned based on a clustering algorithm, which was initially developed for the identification of communities within social networks.
The underlying service definition and many other details of the meet-in-the-middle approach remain unclear, however. While a differentiation of service hierarchies and types is mentioned, its realization in the supporting tool, the EA-Builder, is not further addressed. A quality assertion of the results is missing, but the approach has been tested on the basis of a use case.
Based on an analysis of business processes, needed services are identified in a top-down approach. Existing services are extracted from the code base and the related data structures. By comparing needed and existing services, additionally required services are identified. A so-called tool-based mining supports the bottom-up analysis of code and data fragments. The top-down analysis of the business processes is realized by a combination of interviews and tools.
The approach does not present a concrete definition of services and, besides notes about possible tool support, no tools are mentioned nor does it offer a procedure for matching needed with existing services. While the authors present a case study, their explanations only cover central results and do not document the practical application of the approach at all.
Potential services are identified and opposed to each other in four steps using a top-down as well as a bottom-up approach. Additionally needed services are added in an undefined way. While this contribution describes the identification in a more structured and detailed manner than Zimmerman et al. Optimization methods as well as tool support are not considered. A reference example is not given either. During the identification phase services are defined based on UML activity diagrams. These services have to comply with three previously defined criteria, namely reusability of services, avoidance of redundant implementation of different services as well as loose coupling of services based on well-defined and simple interfaces.
The service identification itself has four subsequent steps: On the basis of an implicit business-oriented service definition, the service identification follows a semi-structured top-down approach. An optimization of determined services is not part of the identification phase and the compliance of the identified services with the previously defined criteria is not validated. While service hierarchies, service dependencies and structures are mentioned, it stays unclear how these aspects affect the service identification.
The whole process is described on the basis of an example from the financial service sector. A supporting tool is not mentioned. As a result, they offer an own approach, which covers the phases of service identification and specification and is integrated into an overall development process model. Services are identified through a top-down decomposition of business processes.
Special attention is placed on an analysis of so-called transfer and visibility potentials of single process steps for business partners. While existing structures and services are taken into account during the identification phase, the dependencies between services are only considered in the specification phase. Service hierarchies are divided into two types, Process and Basic. A differentiation of service types is mentioned, but not integral part of the approach.
Although a structured identification process is propagated, further details about such a process are missing. Moreover, neither a possible optimization nor a supporting tool is mentioned. A usecase demonstrates the practicability of the approach, but does not describe any details of the sub-steps. The approach offers detailed steps and procedures to identify system modules by examining the relations between process activities and data objects in a matrix analysis.
Based on a heuristic optimization procedure, the grouping of process activities is rearranged to minimize the number of shared data objects between the groups of activities. A quality assertion for the results of this optimization is not given and existing system structures cannot be incorporated into the presented approach. Furthermore, it remains unclear how the rearrangement of process activities should be conducted. According to these criteria, services should not only be self-contained with respect to their functionality, but also be independently implementable, installable, and maintainable.
In addition, they should be independent with respect to billing and handling of liability issues. While the criteria are formulated in detail, they do not exceed the level of general recommendations. A structured procedure with concrete work steps to guide the designer is missing completely. The approach can therefore only be characterized as a conceptual framework which might be used to validate identified services.
Based on the Analysis Level Object Model, a business domain model in UML notation, the approach identifies reusable components following a topdown approach. The domain model contains at least object-oriented class diagrams, use cases and sequence or interaction diagrams. Structural and dynamic relationships between the different objects in the domain model are used to compute the Class Relationship Strength. In a first step these relationships are used to identify components through a grouping of classes by applying a hierarchical agglomerative clustering algorithm.
This initial solution is then improved through automated add-, move- or exchange- heuristics, as well as manual interventions. The identification process is II. An evaluation of identification results is not mentioned. Both versions identify components by using algorithms that work on process and data structures of a domain model.
The component identification follows an algorithmic top-down approach, which is supported by specialized tools and part of the Business Component Modelling Process BCMP. Information about data objects, process steps and actors, plus their relationships is mapped onto vertices and edges of a graph. A quality assertion for the resulting solution is not given and legacy structures as well as existing dependencies can only be partly included by integrating them into the domain model.
The BCI-3D tool supports the designer during the identification process, but missing advices for the assignment of weights hamper the application of the proposed method. Whereas all of the specialized service identification approaches were published between and , the more general modularization approaches are mainly older and dated before An overview of the classification results of all approaches is summarized in Table 2. It shows that none of the 9 specialized and the 4 general approaches covers all aspects sufficiently, but rather all of them have their individual strengths and weaknesses.
Below follows a differentiated examination and comparison of the approaches. Classification of service identification approaches. However, an interesting observation is that the authors of eight approaches identify services without any definition of what they try to find. Furthermore, even if a definition is given, it is oftentimes imprecise and therefore handicaps a comparison of the approaches. In the remaining six cases general recommendations are given.
It is noticeable that the older, general modularization approaches are overall more formalized than those originating from the newer SOA discipline. On the other hand, most of the newer approaches are embedded in a development process model and thus better support an integrated design than the older ones.
Generally, the amount of information and sub-steps explained varies noticeably in the evaluated literature, ranging from detailed step-by-step instructions to rather coarse-grained explanations. A low level of detail especially hampers the applicability of an approach. Interestingly, the level of detail is not correlated with the degree of formalization, which one might have expected. Also, not a single one of the approaches uses a bottom-up strategy.
Generally, it seems like the specific service identification approaches tend to consider technical information more often in combination with business domain information than the more general modularization techniques which solely rely on the business domain. One of the explicit service identification methods uses a heuristic to optimize the results [Aier06], whereas the other eight do not implement any optimization.
Existing structures and system dependencies are each taken into account by nine of the approaches. This is usually granted for the meet-in-the-middle approaches, but for the top-down approaches it can only be achieved through an integration of the information into the domain model. Two third of the proposed techniques for service identification care for service hierarchies and include corresponding arrangements in their strategy.
The differentiation of predefined service types is covered in one third of the papers. Service hierarchies and service types are not applicable for the non-service-specific approaches. Quality assertions are so far completely missing for all of the approaches. For the evaluation of the approaches, use cases and plausibility checks are provided in eight cases, but none is evaluated through best practices.
Overall, this does not go far enough to ensure a high quality solution. However, this would be crucial for the further development of the procedures. The former ones can use the results from Table 2 to identify areas requiring further research, improve their own approaches and fill out the blanks. To improve the usability of several methods, the analysis of the supporting measures shows that software tools are needed, especially in the case of optimizing approaches.
Furthermore, the given quality assertions and evaluations are mostly quite rudimental. Additionally, researchers might be able to use combinations of existing service-specific and general approaches for further improvements of the state of the art in service identification. For example, we see a good chance that the BCI approach from Albani et al. Above all, the utilized service definition, the direction of the approach as well as the required model views provide insights whether an approach is suited to support a particular development scenario or not.
In a greenfield software development project, where services can be identified during the early II. But in general the decision for a specific approach should be depending upon whether the overall goal during the identification is to identify reusable services or to primarily create a modular system design. In a scenario where existing software systems have to be modularized or at least to be integrated into the identification of services, meet-in-the-middle approaches should be preferred.
These approaches either start from existing software systems and aggregate implementation classes to form services or at least take existing software structures into account during the identification of services. As the classification shows, an integrated service identification approach, which combines the strengths of the mentioned approaches and is able to cover all depicted scenarios, is not available so far. Therefore, it currently depends on the knowledge of the designer, if a suitable approach is chosen and useful results can be achieved. Our paper thus provides useful insights by identifying and detailing on the state of the art.
By analyzing the classified approaches from a chronological perspective, it becomes furthermore obvious that the approaches related to the older discipline of component-based application development usually possess more structured, algorithm-based and better evaluated procedures compared to the approaches that specifically support the identification of services. It also has to be highlighted that most of the service-specific identification approaches come without an explicit service definition.
For the designer, it therefore often remains unclear which criteria of services are assumed and guaranteed by the approaches during the identification process. In conclusion, service-specific approaches hence appear to be in a comparatively premature state. Additional research effort is therefore required to further advance the state of the art and support an engineering approach to identify services.
New PDF release: Möglichkeiten und Grenzen zum Customer Relationship - vacances aux Book Archive
The discussion was based on a classification scheme that contains various characteristics of service identification approaches as distinguishing factors and has been specifically developed to compare existent as well as future developments. The assembled characteristics were initially based on results from research focusing on the conception of systematic design methods in general and then refined to characterize service identification approaches. We used the resulting classification scheme to compare various service identification approaches and to reveal differences in their conceptual design as discussed in section 4.
A stepwise refinement, which might become necessary during the service identification process, is therefore not supported by all approaches. Approaches distinguishing so-called Entity Data and Task Process Services are likely to promote different solutions than those which do not build upon such a predefined distinction of service types. This distinction has to be taken into account when a certain approach is selected. Significant further research is necessary to answer the question if the existing approaches for service identification can be further developed into mature methods with more formalized and detailed procedures.
To realize the aspired systematic service identification as part of an engineering process, existing approaches will eventually have to be enhanced in various aspects. In this context, especially the usage of optimization methods and procedures, which allow at least an estimation of the solution quality, has to be considered.
It appears to be characteristic that the identification of services does not build upon existing, more mature approaches from the closely related and older component-based software engineering discipline. Instead, it seems as if research has started anew with the introduction of SOAs.
A more detailed examination reveals that many modularization approaches, which were developed to identify business components, as defined by Cheesman and Daniels [ChDa01], could well be used for the design of service-oriented architectures. A synergetic examination of these two disciplines could hence significantly accelerate the development of mature methods for service identification. References [Aier06] Aier, Stephan: Interoperability of Enterprise Software and Applications.
Springer Verlag, Geneva, Switzerland, , pp. IBM developerWorks Web services zone, http: The Rise of Web Service Ecosystems. IT Professional 8 5, pp. George Braziller, New York, Wirtschaftsinformatik 50 3, pp. A research manifesto for services science. Communications of the ACM 49 7, pp.
Object-Oriented Modelling with Syntropy. International Business Machines Corporation, Graphs, networks, and algorithms. Informatik Spektrum 26 2, pp. Prentice Hall, London, Debugging Techniques in Large Systems. Prentice Hall, , pp. Proceedings of Dagstuhl Seminar. Communications of the ACM 15 12, pp. Enterprise Services Design Guide. International Journal of Information Technology 2 4, pp. Analyzing the past to prepare for the future: Writing a literature review.
Dissertation_Sebastian_Kloeckner.pdf - OPUS Augsburg - Universität
MIS Quarterly 26 2, pp. Wirtschaftsinformatik 49 4, pp. Auf Basis dieser Merkmale wird dann ein entsprechendes Architekturkonzept zur Integration verteilter Unternehmensdaten abgeleitet. Business Process Management Workshops: Springer, Berlin, Heidelberg , S. Mit dem Einsatz spezialisierter Anwendungssysteme, bspw. In order to support not only intra- but also inter-organizational business processes along the value network, the systems used in the single network nodes need to be integrated.
This implies that the IT application systems of customers, partners and suppliers need to be integrated into an inter-organizational system in order to allow automated data interchange in the value network. The integration includes different functional areas like service, marketing, sales, procurement, production, distribution and waste disposal. Only the integration of all these functional areas, which are directly involved in the value creation, allows a realization of a transparent and continuous supply network.
Thanks to the deployment of enterprise resource planning ERP system internal integration is already on a high level and at the same time precondition for the realization of inter-organizational integration, which contains a potential by far larger than that of intra-organizational integration []. Nonetheless, the realization of inter-organizational integration in the business world is not as far as science has explored it. This is mainly caused by the lack of standards and absence of adequate information system architectures, which contain the construction plan of the information system — in terms of a specification and documentation of its components as well as their relationships — as well as the rules of construction for the creation of such a construction plan [7].
As an inter-organizational integration has an enormous influence on the design of information systems and as preliminary and downstream enterprises in the supply network have to be integrated, the creation of an inter-organizational information system architecture is necessary in order to ensure a continuous exchange of information between the involved partners throughout the whole value network. But before such an inter-organizational integration becomes possible, a corresponding intra-organizational integration has to be achieved. These two concepts cover most of the functional areas involved directly in value creation and are therefore adequate as basis for the development of an integrative information system architecture [9].
CRM creates the connection with the customer within the functional areas of service, sales and marketing, while SCM creates the connection with the suppliers, partners and customers through the functional areas procurement, production, logistics and waste disposal. In doing so the following research question will be answered: Why is an intra-organizational integration necessary in order to achieve inter-organizational integration?
In what way does a business components-based information system architecture contribute to the interoperability in a value network? What types of business components are relevant for the interoperability and how do the components need to be composed in an inter-organizational system? Therefore, in section 2 the state of the art of inter-organizational systems is illustrated, showing the necessity of integrating CRM and SCM in an enterprise in order to provide an adequate basis for the development of inter-organizational systems.
In section 3 an integrative information systems architecture for CRM and SCM is derived in order to ensure interoperability. Section 4 explains the integrative architecture by means of two examples. Concluding remarks and future work can be found in section 5. As far as only the relationship between the supplier and the customer is taken into account this kind of e-collaboration is adequate. But as actual real-world examples show the limitation of focus to only one relationship in the whole supply network as area of interest is not sufficient.
While the considered relationship between supplier and customer performs well, problems at preliminary stages of the supply network might cause extensive interruptions of production processes downstream. Demand driven value networks [10, 11] do not just only take the relationship from the OEM to the subsequent tier into account, but the whole supply network with several tiers. As shown in Fig. The resulting information allows each supplier to judge if he and his preliminary suppliers are able to accomplish the potential customer order.
But the assumption that every node is able to process an incoming request for quotation and convert it into outgoing requests for quotation for his own preliminary suppliers is not valid in most cases. Actual state of interconnection of internal and external systems In fact, most often the internal systems of a company are coupled by standard interfaces or customized adapters. As the interfaces and adapters do not fit exactly with the coupled systems, the systems cannot communicate with each other correctly resulting in a loss of information or functionality.
The loss of information can be caused by several reasons: As a result, the reliability and actuality of data cannot be assured making it impossible to determine if an order or request can be fulfilled or not. Second, as structures and semantics of data can vary between the systems, a matching of the ontologies can be difficult or sometimes even impossible [].
The loss of functionality can also be caused by several reasons: Consequently, this function cannot be called by an external system. Second, even if a public interface or API is offered, the signature of the interface may not be suitable for another system as additional information would be needed. Accordingly, even if the function could be called by an external system, it would not operate correctly as not all input parameters may be available [16, 17].
While a direct connection between a CRM system and a SCM system would be favorable, in most cases such a direct interconnection does not exist. Therefore, the data exchange between those systems inside an enterprise is very limited, while it would be the precondition for successful interconnection of the whole supply network. In order to solve this problem, direct connections between these systems have to be established. Direct interconnection does allow to reciprocally access the data and to use the functionality of the independent systems.
The publication of interfaces of subsystems of a concerned system allows additional enhancement of the interconnection of the systems. Consequently, each subsystem could be interconnected making it possible to make use of further functionality. But even if all interfaces are published and available to external systems, the problem of redundant data, concurrent updates and different data structures still exists as both systems contain independent data management functionality. For solving this problem an integrated III.
Within the integration context some functions of one business area are assigned to a business component [22, 23] containing additional functions of different business areas and providing the functionality to the outside world over well defined services [23] following the idea of a service oriented architecture. Under an integrative perspective business components are no longer strictly predetermined by membership in a certain business area, but rather composed in a way that relationships of information objects are optimized.
The goal of the composition is to maximize the exchange of information within a business component and minimize the exchange of information between the business components while simultaneously avoiding that the technical purpose of the business III. The rearrangement of functionality does not only allow direct access to functions and information objects of other business areas.
In addition, due to the incorporation in one integrative ISA, other problems like data redundancy or data matching issues are solved, because of a coordinated data management. Consequently, this integrative information system architecture allows the inter-organizational integration of organizations involved in the supply network and therefore a continuous exchange of information throughout the whole supply network.
- Gregory of Nazianzus (The Early Church Fathers).
- THE VINEYARDS OF ALLEGRETTI!
- Dissertation_Sebastian_Kloeckner.pdf - OPUS Augsburg - Universität.
- Entscheidungsfindung im Cloud Computing Konzeption und Analyse eines Modells zur Anbieterauswahl.
- The Smelliest Day at the Zoo.
Nonetheless, an interconnection or integration of such a system can be achieved in the same way as it was shown for SCM and CRM. For the development of the integrative information system architecture the Business Component Identification BCI -Method [24, 25] has been used, which is based upon the Business System Planning BSP [26] method and which has been modified for the field of business components identification.
The basis for the BCI method is a well elaborated domain analysis. The BCI method takes as input the business tasks of a specific domain, as e. In a first step a matrix is built defining the relationships between the single business tasks and the informational data. In changing the order of data and of business tasks according to some metrics defined — e. The result of the BCI is an abstract business component model with some already defined dependencies between components [24, 27, 28].
Due to display reasons an illustration of all identified business components is not feasible, whereas the relevant components for the examples illustrated in the next section are described in detail in section 4. Additionally, one further component is included: Its purpose is to coordinate the communicating between the other components.
This allows simple adjustments of the communication channels if the services of one or more components are altered. The first example process shows a request for quotation, which can be satisfied without additional requests to preliminary suppliers. The second example extends the first one by assuming that the organization does not have all required parts on stock.
Consequently, preliminary suppliers have to be involved in order to answer the request of the customer correctly. The preliminary suppliers themselves also have to contact their preliminary suppliers for evaluating the availability of the required parts. Due to complexity reasons only three of the seven examined functional areas of the integrated ISA are included. Both examples show the informational relationships of the business components involved.
As already mentioned above, the orchestration component is also included. Additionally, a communication component, which is responsible for the communication with external systems, is incorporated for completeness reasons. Its functionality is not explained further in this paper as it does not have a direct influence on the ISA, but on the inter-organizational communication. Prior to the illustration of the example processes, the business components of the example processes are explained. From all the business components of the component model presented in section 3 only the 6 components, which are relevant for the example processes, are described next.
Customer Order The business component customer order contains all necessary functionality for further processing of a customer quote. With regard to the example process the order execution planning and the delivery date confirmation are of particular importance. Additionally, all basic agreements of the customer order are arranged and submitted to the customer. Material Requirements The business component material requirements prepares the on-time allocation of all required materials in regard to type, quality and amount for the production process.
Capacity Planning The functions of the business component capacity planning include design of capacity capability, determination of the capacity demand, deployment of staff, comparison of capacity demand and availability, scheduling of capacity and planning of machine allocation sequence. Stock of Inventory The business component stock of inventory contains the functionality which is associated with stock movements. Within others, this affects the inventory management, which is the link between demand and order planning. Purchasing Transaction The business component purchasing transaction includes all functions which are needed for the transaction-oriented part of the procurement initiation.
This involves acceptance and transmission of requirement requests, requests for offers, which precede the registration and evaluation of requirements. Manufacturing lead time scheduling The business component manufacturing lead time scheduling contains all relevant functionality for successful scheduling like determination of process steps and operations or definitions of process and administrative times.
By creation of task schedules, determination of lead times and evaluation of possible lead time reductions preliminary starting and ending times for the different process can be generated. A customer asks the OEM for a customized product. Besides the price and the configuration of the product, the sales employee wants to tell the customer the delivery date as it is one of the crucial factors for his purchase decision.
The determination of the delivery date requires access to all functional areas involved.