지식집약 서비스를 위한 휴먼 인터액션의 분산화된 관리 방법론

Abstract

Knowledge intensive service activities provide an opportunity for companies to increase competitiveness and accelerate innovation by bringing valuable external knowledge to them. A knowledge intensive service is a specialized business service that aims to create value-added activities and customized solutions to meet client’s needs by relying heavily on knowledge work. As the knowledge work is high level professional work which contributes to the satisfaction of clients’ needs, knowledge workers are the most valuable resources for service-providing companies. The workers generally undertake complex operations of an intellectual nature to develop and deliver the services. Therefore, systematical management of complex service processes is the most important objective that the service-providing companies have. Traditional Business Process Management (BPM) methodologies and solutions have been applied for this purpose as they have allowed the service-providing companies to deal with a wide range of the process management. However, most of the concerns that they take care of are mechanistic system-to-system scenarios with pre-defined workflows. The way humans work in complex service processes is not quite the same as the way mechanistic systems do, since the systems simply follow a pre-defined sequence of activities one after the other. Humans inherently do what they like and what they think necessary. Consequently, it is necessary to develop a way to manage the service processes in a human-oriented way. To model and manage the human processes, Human Interaction Management (HIM) lays a role concept at its heart since the role concept is generally regarded to be compatible with general human work behavior. There are many different role concepts applied in different systems. Among them, roles highly relevant to human collaborative systems are discussed in HIM. In this respect, a role can be defined as a set of prescriptions describing how people should behave with responsibilities required to carry out specific activities. Thus, through clearly defined roles in human processes, humans can be well informed about what objects they can access with specific rights and which people they can manage or communicate with, and ultimately they can be facilitated in accomplishing their jobs meaningfully and efficiently. Although HIM suggests a complete theory about managing human processes based on the role concept, it does not provide concrete methodologies for applying its principles to the modeling and management of the knowledge intensive service processes. In order to ensure the successful management of the human processes, the following problems should be addressed: 1) overwhelmed messages in diverse human interactions, 2) lack of a mechanism for dynamic process change management, and 3) lack of a mechanism for dynamic role creation and assignment. This research focuses on helping people interact better with each other by presenting several methodologies which can provide a systematic guideline as to how they work together. For this purpose, this research is performed as the following three issues.The first issue, “A State-Driven Modeling Approach to Human Interactions,” deals with the method of how the human interactions can be clearly described and effectively managed in a human-oriented way. Moreover, it addresses how to guide the participants with the proper information like what they can do at a certain specific time. The second issue, “A Decentralized Approach to Dynamic Changes of Human Interactions,” deals with the method of how the interaction models can be dynamically and continuously changed by the participants. The change should not be a one time thing but happen continually through an agreement among the involved participants. To manage the model change in a decentralized way, this issue presents a procedural framework for the dynamic changes of human interactions. The approach enables the participants to monitor every essential element of the interaction, and subsequently to find various potential problems that are currently occurring in the interaction. Moreover, the approach makes the reuse of interaction models feasible by streamlining the retrieval of similar models from the model repository using the measurement of semantic similarity among the goal descriptions of the models. The third issue, “An Ontology-Based Approach to Role Creation and Assignment,” deals with the method of how new roles can be created and assigned to appropriate humans at execution time. The proposed methodology specifies the information about roles and humans ontologically and identifies appropriate candidate humans for a certain role based on the ontological specifications. Moreover, the participants can be provided appropriate candidate humans when carrying out the assignment task from the methodology. Therefore, they can easily create new roles and introduce other people into the ongoing processes during the process enactment.Through these issues, this research provides the following contributions: 1) providing a synthesized architecture for dynamic modeling of human interactions, 2) concretizing methodologies in the synthesized architecture, 3) providing a basis for implementing human process management system, and 4) facilitating proactive collaboration between human participants.