LEMO Project

This is a proposal for a student project at the School of Business of the University of Applied Sciences Northwestern Switzerland FHNW. Important: The following content has a hypothetical character and will be evaluated within several master theses.

Transcript

1. Andreas Martin - FHNW Linked Enterprise Models and Objects Linked Enterprise Models and Objects (LEMO) An integrated way in enterprise modelling: Interlinking enterprise models and enterprise information objects. Linked Enterprise Models and Objects (LEMO) 1 Linked Enterprise Models Linked Enterprise Data & Information Objects
2. Andreas Martin - FHNW The information challenge  In the current world enterprise are confronted with different information challenges. Starting with the information technology perspective, enterprises are confronted with multiple systems and technologies. This makes it difficult to get a comprehensive and consistent view of all information systems that are running. Apart from the information technology diversification itself, good data quality and specification are absolutely important for decision making and expectation meeting. Finally the information flow, it is all about communication and speaking the same language in an enterprise. Linked Enterprise Models and Objects (LEMO) 2
3. Andreas Martin - FHNW Enterprise models  Enterprise models are an excellent way to explain, constrain, guide and predict things in a company. They contain valuable knowledge about the enterprise itself and provide adequate representations to different stakeholders. Good (enterprise) models are correct, relevant, economical efficient, clear, comparable and systematically designed (Becker et. al.). Enterprise models are a brilliant tool to understand what already exists in a company and what will be in future - that’s conversation by enterprise models. They do an excellent job when designing and shaping the future of an enterprise - guidance by enterprise models. And with the usage of enterprise models it is possible to verify and asses the current situation of the enterprise itself - that's compliance by enterprise models. Linked Enterprise Models and Objects (LEMO) 3
4. Andreas Martin - FHNW Linked enterprise models  Enterprise models can only fulfil these high expectations when they are made (1) explicit, (2) linked to other models and based on a (3) common shared understanding. To produce consistent, relevant and clear enterprise models they have to be made explicit. Otherwise "[…] you are making assumptions about any […] model that you have not made explicit and those assumptions may be right… or they may be wrong" (John A. Zachman, 2012). Consistent enterprise models can't exist in an isolated manner - the reality is not isolated. It is extremely rarely the case that a business process model has no relationship to an information model, document model or an organisational chart. It is obvious that enterprise models are linked at least implicitly even they not build using the same modelling language. In essence, enterprise modelling should be about people, audiences and divisions talking to each other. To achieve this, people need a common shared understanding about the enterprise. They need to speak the same "language" using a common vocabulary. As a result, enterprise models should contain and reuse the enterprise specific common vocabulary. Linked Enterprise Models and Objects (LEMO) 4
5. Andreas Martin - FHNW Why are linked enterprise models important to business?  First of all enterprise models are a great method for an open and efficient communication over different divisions and stakeholders in an enterprise. Linked enterprise models enable the reusability of existing and shared concepts and circumstances in other models. This gives the possibility that an enterprise do not need to start from scratch again and again when modelling new perspectives. Apart from that, linked enterprise models are an excellent may to enhance the transparency within an enterprise. Linked Enterprise Models and Objects (LEMO) 5
6. Andreas Martin - FHNW Linked Enterprise Models & Enterprise Vocabularies  A common enterprise vocabulary enhances the reusability of common terms in enterprise models. An enterprise ontology ensures that the enterprise models are consistent.  “[…] vocabularies define the concepts and relationships (also referred to as “terms”) used to describe and represent an area of concern” (W3C).  “An ontology is a formal, explicit specification of a shared conceptualisation” [34, p. 184].  “There is no clear division between what is referred to as vocabularies and ontologies. […] The trend is to use the word ontology for more complex, and possibly quite formal collection of terms […]” (W3C).  “A controlled vocabulary is a list of terms that have been enumerated explicitly.” In a ideal situation all terms “[…] should have an unambiguous, non-redundant definition” (Pidcock 2003).  “A formal ontology is a controlled vocabulary expressed in an ontology representation language” (Pidcock 2003). Linked Enterprise Models and Objects (LEMO) 6
7. Andreas Martin - FHNW Linked Enterprise Models & Enterprise Ontologies  “The main purpose of an enterprise ontology is to promote the common understanding between people across enterprises, as well as to serve as a communication medium between people and applications, and between different applications” [35, p. 273]. Linked Enterprise Models and Objects (LEMO) 7
8. Andreas Martin - FHNW Linked Enterprise Models & Enterprise Ontologies including Enterprise Architectures  “Enterprise Architectures (AE) are a way to model the relevant aspects of an enterprise and interdependencies between business and information systems” (Martin et. al., 2013).  According to Zachman, an Enterprise Architecture Framework is an ontology and a “structured set of essential components of an object for which explicit expressions is necessary and perhaps even mandatory for creating, operating, and changing the object […]” (Zachman, 2008) e.g. an enterprise.  The Zachman Framework is not formalized in an ontology representation language till now – but it should be. The FHNW has introduced a formalized enterprise ontology called ArchiMEO, which is based on enterprise architecture framework ArchiMate. Linked Enterprise Models and Objects (LEMO) 8
9. Andreas Martin - FHNW Linked Enterprise Models - Application Enterprise vocabulary / ontology: can improve the communication between people inside an enterprise.  Information queries e.g.  Information demand: “Who need which kind of information?”  Model queries e.g.  Model correctness: “Are our enterprise models consistent?”  Model relevance: “Do we need that enterprise model?” Linked Enterprise Models and Objects (LEMO) 9
10. Andreas Martin - FHNW Making more out of it: Linked Enterprise Models & Objects  Information integration and data analysis is a hot topic in enterprises nowadays. This can be observed that the term Big Data is used inflationary. Detractors argue that: "The big thing about big data is that there are too much data“. True or not - important is that existing data of an enterprise is used in an intelligent way, which refers us to the term “smart data”. In fact the term “smart” does have different meanings. “Smart” can mean “intelligent” and (apart from other meanings) it can stand for “sharp pain”. Well, both meanings can be true for enterprise data. Data can be inconsistent, missing, false, etc. and important it is often the case that data is not based on a enterprise specific common shared understanding. That makes it difficult to reuse and integrate – “that’s indeed a sharp pain”. As a hypothesis, a possible solution to that problem would be to integrate and interlink all relevant data or information objects using the enterprise vocabulary formalized as enterprise ontology. The enterprise ontology can act as a integration schema. Linked Enterprise Models and Objects (LEMO) 10
11. Andreas Martin - FHNW Linked Enterprise Models & Objects - Application  Enterprise over-view: dynamic reports for different stakeholders (e.g. management cockpit).  Information queries e.g.  Information redundancy: “Do we have redundant information?”  Information inconsistency: “Do we have data, which is not consistent according to an enterprise vocabulary?”  Model queries e.g.  Model efficiency: “Does your enterprise work as expected?” Linked Enterprise Models and Objects (LEMO) 11
12. Andreas Martin - FHNW The HSW-LEMO Project  The goal of this joint research project is to evaluate the hypothesis of “Linked Enterprise Models and Objects (LEMO)”.  It is designed as a joint research project of different master students working on their own research questions.  The variety of topics is within the LEMO. Nevertheless, there are business oriented and technical oriented topics available.  Apart from the literature review, the project starts with the creation of the application scenario.  The application scenario will be created during the creation of a real world enterprise architecture. Linked Enterprise Models and Objects (LEMO) 12
13. Andreas Martin - FHNW The HSW-LEMO Project  The students are invited to create an enterprise architecture for the School of Business FHNW during research proposal phase.  This gives a unique opportunity to put the enterprise architecture methodology into practice and have access to real world data. Apart from that, our experts in enterprise architecture (Knut Hinkelmann and others) will coach the students. Linked Enterprise Models and Objects (LEMO) 13
14. Andreas Martin - FHNW Overview of LEMO and Research Topics (1) LEMO Method (1 MSc Thesis) (2) LEMO Cycle (1-3 MSc Thesis) (3) LEMO Incorporation (1-3 MSc Thesis) (4) LEMO Cockpit (1 MSc Thesis) (4) LEMO Browser (1 MSc Thesis) (5) LEMO Integration & Change (1-2 MSc Thesis) Linked Enterprise Models and Objects (LEMO) 14 (1) (2) (3) (3) (4) (5) (5)
15. Andreas Martin - FHNW Research topics of the LEMO project 1. LEMO Method (1 MSc Thesis): Implementation of LEMO in an enterprise. Introduction of a Development Method. 2. LEMO Cycle (1-3 MSc Thesis): Evaluation and exemplary implementation of Enterprise Model & Ontology Roundtrip approaches: Semantic lifting, MDA- based approach, etc. 3. LEMO Incorporation (1-3 MSc Thesis): Incorporate, embrace and include other Enterprise Models and Enterprise Architecture Frameworks: Zachman Framework, ArchiMate 2.0, BPMN, etc. 4. LEMO Cockpit (1 MSc Thesis): An enterprise cockpit and query analyser presenting the telemetry of an enterprise. 5. LEMO Browser (1 MSc Thesis): Enterprise vocabulary browser and visualisation. Cognitively adequate visual representation of an enterprise ontology for ontology engineers and domain experts. 6. LEMO Integration & Change (1-2 MSc Thesis): Integration Best Practices: e.g. integration of data from relational databases, DMS, etc. / integration of the LEMO approach into an existing IT environment Change Management and Schema changes, etc. Linked Enterprise Models and Objects (LEMO) 15
16. Andreas Martin - FHNW LEMO Method  Teaser: Design a new generalized, step-wised an iterative method for enterprise modelling.  When introducing an enterprise architecture in an enterprise a step-wised development method is used. An example of such an development method is the enterprise architecture framework TOGAF, which is seen as “[…] reliable, proven method for developing an […] enterprise architecture […]”.  Problem: Up to now, there is no all-embracing development method available, which combines the enterprise architecture and enterprise ontology perspectives.  Objectives: The goal of this work is to introduce a generalized, step-wised an iterative method for developing an enterprise architecture using an enterprise architecture language and the enterprise ontology ArchiMEO. The new approach can be developed and evaluated using the School of Business as a real world scenario.  This research problem can be processed by one student.  Supervisors: Barbara Thönssen and Andreas Martin Linked Enterprise Models and Objects (LEMO) 16 Project Characteristics Business Technology Information Systems Enterprise Modelling W3C Semantic Web Enterprise Software Eng. Enterprise Architecture F. Model Driven Engineering User Centric Design
17. Andreas Martin - FHNW LEMO Cycle  Teaser: Introduce an innovative cyclic roundtrip approach, which enables to keep enterprise models with an enterprise ontology in-sync.  Enterprise models are usually created using sophisticated specialized modelling and CASE tools (such as Enterprise Architect by Sparx Systems, Agilian by Visual Paradigm, Adonis by BOC, ARIS by IDS Scheer, etc.). The OMG introduced the term model driven architecture (MDA) that specifies a way for transforming MOF- based modelling languages into other MOF-based languages.  Problem: To reuse existing knowledge captured in enterprise models, a time consuming handmade transformation is needed, which can lead to inconsistency too.  Objectives: The objective of this research work is to evaluate and introduce an cyclic roundtrip (export, import) approach between a given standardized model (e.g. ArchiMate or/and BPMN, etc.) and the enterprise ontology ArchiMEO). Potential methods that would be worth considering are: Model Driven Architecture (MDA) and semantic lifting.  This research problem can be processed by up to three students individually.  Supervisors: Knut Hinkelmann and Andreas Martin Linked Enterprise Models and Objects (LEMO) 17 Project Characteristics Business Technology Information Systems Enterprise Modelling W3C Semantic Web Enterprise Software Eng. Enterprise Architecture F. Model Driven Engineering User Centric Design
18. Andreas Martin - FHNW LEMO Incorporation  Teaser: Work on John Zachman’s vision of an Enterprise Ontology - interlink and formalize Enterprise Architecture Frameworks.  The Zachman Framework is one of the most prominent enterprise architecture frameworks nowadays. The “father of enterprise architecture” John Zachman is describing its own architecture framework as an ontology – “the Enterprise Ontology”. Unfortunately, the Zachman Framework is not available as formalized enterprise ontology. Apart from that, the well known enterprise architecture framework ArchiMate is available as enterprise ontology in version 1. This formalized version of ArchiMate is called ArchiMEO and developed by the FHNW – now it is time to enhance ArchiMEO with other enterprise models (e.g. BPMN, etc.) and update it to version 2.  Problem: There exists several enterprise architecture frameworks and enterprise models, which can’t reused in a formalized way using semantic technologies.  Objectives: Incorporate, embrace and include other Enterprise Architecture Frameworks. Suggestions are the Zachman Framework, ArchiMate 2.0 and others.  This research problem can be processed by up to three students individually.  Supervisors: Barbara Thönssen and Andreas Martin Linked Enterprise Models and Objects (LEMO) 18 Project Characteristics Business Technology Information Systems Enterprise Modelling W3C Semantic Web Enterprise Software Eng. Enterprise Architecture F. Model Driven Engineering User Centric Design
19. Andreas Martin - FHNW LEMO Cockpit  Teaser: Develop an enterprise cockpit presenting the telemetry of an enterprise.  Management cockpits are important tools to get an overview of the enterprise. These cockpits shows the current data of an enterprise in a compact and consolidated form. As a hypothesis, a management cockpit using semantic technologies can extend and enhance current management cockpit approaches significantly.  Apart from that an enterprise ontology can be used to run enterprise queries. As a hypothesis, these queries can be used to get exact information about every aspect of the enterprise.  This research problem can be processed by one student.  Supervisors: Knut Hinkelmann and Andreas Martin Linked Enterprise Models and Objects (LEMO) 19 Project Characteristics Business Technology Information Systems Enterprise Modelling W3C Semantic Web Enterprise Software Eng. Enterprise Architecture F. Model Driven Engineering User Centric Design
20. Andreas Martin - FHNW LEMO Browser  Teaser: Cognitively adequate visual representation of an enterprise ontology that enables domain experts, modelling experts and ontology engineers to get a quick overview of the enterprise vocabulary and existing models.  To create consistent enterprise models using a enterprise vocabulary it is required to get straightforward and uncomplicated overview of the enterprise ontology. As a hypothesis, a cognitively adequate visual representation of an enterprise ontology can improve the enterprise modelling significantly.  Objectives: The objective if this research work is to evaluate existing visualisation approaches and propose an cognitively adequate visual representation. It is possible to use the FHNW usability lab for evaluation purposes.  This research problem can be processed by one student.  Supervisors: Knut Hinkelmann and Andreas Martin Linked Enterprise Models and Objects (LEMO) 20 Project Characteristics Business Technology Information Systems Enterprise Modelling W3C Semantic Web Enterprise Software Eng. Enterprise Architecture F. Model Driven Engineering User Centric Design
21. Andreas Martin - FHNW LEMO Integration and Change  Teaser: Enrich enterprise IT with semantic information of an enterprise ontology using a mediator based integration scenario.  An enterprise knowledge base containing enterprise architecture and enterprise model information is usually not isolated. In an enterprise there exists typically several information sources (like databases, enterprise applications, etc.). When integrating information sources and using an ontology in a productive environment the change management plays an important role. Change Management deals with alignment of staging and production environment, comparison and synchronization of the schema and data including incremental schema changes.  Problem: There exists no comprehensive approach for integrating different information sources into an enterprise ontology based triple-store including change management.  Objectives: The objective if this research work is to evaluate different integration approaches and to implement a exemplary scenario. Evaluation of a small-scale semantic web tool stack. Development best practices and a reference system architecture. The evaluation of a schema change lifecycle.  This research problem can be processed by up to two students individually.  Supervisors: Barbara Thönssen and Andreas Martin Linked Enterprise Models and Objects (LEMO) 21 Project Characteristics Business Technology Information Systems Enterprise Modelling W3C Semantic Web Enterprise Software Eng. Enterprise Architecture F. Model Driven Engineering User Centric Design