PROCESS KNOWLEDGE MANAGEMENT 27

• Inputs and Outputs. One place knowledge comes into play in a process is as a production input, which is to be operated upon and transformed into an output.

• Controllers. Another place where knowledge comes into play in a process is in the form of controls. In this case, knowledge can be regarded to be "embedded"

in the process.

• Processors. The "Processor" performs the actions needed to produce a result from the process. This kind of knowledge is "encoded" in the process.

• Design. A lot of knowledge is "embedded" in the processes in the form of spec-ifications for the outputs, the inputs, the routines and the requirements. The knowledge in the process makes itself felt through what is done, when, how, by whom and to what standards.

The first three answers concern the process automation and execution. The last one is associated with process models in the analysis and design phases. Our research objects are business process models at the conceptual level. In [122], Olivé visions a goal of automating IS building by a conceptual schema-centric development (CSCD).

A conceptual schema provides general knowledge about the IS domain and about the functions the IS has to perform. With such a vision, the main purpose of the activity of business process modeling is to elicit the process knowledge of the corresponding IS.

Hence, we take the fourth answer to specify that the process knowledge in our research context is embedded in business process models. This statement is also reflected in [79]

with "process models are carriers of process knowledge, knowledge of how to do things".

2.6.2 Knowledge representation

Nonaka and Takeuchi [114] argued that a successful knowledge management program needs to, on the one hand, convert internalized tacit knowledge into explicit codified knowledge in order to share it, but also on the other hand for individuals and groups to internalize and make personally meaningful codified knowledge once it is retrieved from the knowledge management system. A crucial fact is knowledge representation.

Although process knowledge is embedded in process models, it does not mean all process models are process knowledge, that is, not every process modeling representa-tion is knowledge representarepresenta-tion. Knowledge representarepresenta-tion can be various in terms of the distinct roles it plays, each crucial to the task at hand [22]:

• A knowledge representation (KR) is most fundamentally a surrogate, a substitute for the thing itself, used to enable an entity to determine consequences by thinking rather than acting, i.e., by reasoning about the world rather than taking action in it.

• It is a set of ontological commitments, i.e., an answer to the question: In what terms should we think about the world?

• It is a fragmentary theory of intelligent reasoning, expressed in terms of three components: (i) the representation’s fundamental conception of intelligent rea-soning; (ii) the set of inferences the representation sanctions; and (iii) the set of inferences it recommends.

28 CHAPTER 2. PROBLEM SETTING

• It is a medium for pragmatically efficient computation, i.e., the computational environment in which thinking is accomplished.

• It is a medium of human expression, i.e., a language in which we say things about the world.

The conclusions of the research in [22] are drawn all five roles matter; representation and reasoning are interwinded; different representations are often combined; they have formal equivalence. Various artificial languages and notations have been proposed for representing knowledge, such as CycL [20], IKL [50], KIF [29], Loom [97], OWL [195]

and KM [71]. They are typically based on logic and mathematics, and have easily parsable grammars to ease machine processing.

2.6.3 Knowledge management activities associated with process mod-els

Knowledge Management (KM) is an approach to discovering, capturing, and reusing both tacit (in people’s heads) and explicit (digital or paper based) knowledge as well as the cultural and technological means of enabling the knowledge management process to be successful. Usually, the knowledge management covers both instance level and type level knowledge — learning from instances and abstracting them to reusable knowledge.

In a KM system, knowledge processes essentially revolve around the following steps [174]:

Figure 2.6: Knowlege Process [174]

2.7. SUMMARY 29

• Creation or import. The contents need to be created or converted so that they fit the conventions of the company.

• Capture. Knowledge items have to be captured in order to determine their im-portance and how they mesh with the company’s vocabulary conventions.

• Retrieval and access. This step satisfies the searches and queries for knowledge by the knowledge worker.

• Use. The knowledge worker will not only recall knowledge items, but will process them for further use.

The steps are illustrated in Figure 2.6. In this thesis, the scope of the knowledge management is only limited on type level process knowledge – enterprise/business pro-cess models. Therefore, in the context of our propro-cess knowledge management, propro-cess modeling in knowledge representation language is concerned as knowledge creation.

Importing knowledge can be the transformation of conventional process models in the knowledge representations. Knowledge capture is to capture the essential contents in process models through annotation techniques by creating metadata conforming to ontologies. Process knowledge retrieval and access can be conducted through a conven-tional query or search tools, or by applying the ontology and the inference mechanism to derive further views of process knowledge. The possible uses of process knowledge include analysis of existing process models, reusing the legacy models to create new process models, integrating systems based on the process descriptions, etc.

2.7 Summary

This chapter has outlined the context of this research work. We have introduced the theoretical and technical setting relevant to our research topic, such as modeling the-ories and methodologies, the Semantic Web, ontology, semantic annotation, business process model and knowledge management. Main points are summarized as follows.

• From the discussion of modeling theory and methodology, we has scoped that our research area is at conceptual modeling level.

• Ontology as the key enabling technology for the Semantic Web provides an explicit representation of a shared conceptualization.

• Ontology and conceptual model share certain common grounds, which indicates the potential links between them, e.g. usage combination and technology benefits from each other.

• In this work, a concept modeling language – RML is used to analyze meta-models and ontologies with graphical notations; while, an ontology modeling language – OWL is applied to enable ontology machine-interpretable.

• Semantic annotation is an approach to achieve semantic interoperability of het-erogeneous information resources making use of ontology.

30 CHAPTER 2. PROBLEM SETTING

• Process knowledge is embedded in process models whose semantics might be represented heterogeneously.

• Semantic annotation can be applied in capturing and representing process knowl-edge to facilitate process knowlknowl-edge management such as the retrieval and reuse of heterogeneous process models.

After having established the key areas of our research in this chapter, we will con-tinue to detail survey of state of the art in the these areas, namely, process modeling languages, process ontologies, goal modeling and annotation approaches.

Chapter 3