Research Statement
Cooperation is one of the major reasons why organizations are established. People establish an organization if they can not achieve a goal as a single but they think it is achievable by forming a group. Cooperation is the most powerful tool that has enabled human has been survived for such a long time. Many cooperation mechanisms have been evolved along with human history and many of them are still widely used in modern organizations. My major research interests are focused on the novel design of cooperation mechanisms in organizations such as supply chains and business processes. My research takes two well-known approaches in management science: design science and behavioural science. As a design scientist, I've been proposing novel design of cooperation mechanisms like workflow management systems. As a behavioural scientist, I'm interested in finding principles and theories for effective cooperation among business actors within organizational context. My major methodology to derive such theories and principles is a distributed computer simulation that mimics the structure and dynamics of formal or informal human groups in organizations. The derived theories or principles are expected to be used in the implementation of more effective cooperation mechanisms in today’s organizations.
· Change management of business rules in business process automation (1996 – 1999)
My Ph D topic aimed to propose a novel design for sustainable cooperation among business actors against frequent changes on business rules. In particular, I integrated a rule-based system within a workflow management system to effectively manage changes on business rules that guide the execution of organizational business processes. KWM was proposed as a new type of workflow model which enables modelling business processes based on IF-THEN style rules. K-WFMS was implemented to execute KWM for the automation of business processes. The successful deployment of K-WFMS at KAIST won the Workflow Excellence Award in 1998.
· Cooperation among customers in high technology market (1999 – 2003)
During my post doctor period, we worked on the macro-level analysis of customer behaviour in eBusiness environment. We were particularly interested in network externality (a kind of cooperation mechanism of customers) that affects the decision making of consumers for their purchase of products. We proved that the difference on the characteristic of consumers in a market determines the degree of impact of network externality in a network economy through a simulation-based analysis.
· Mobile business processes (2001 – 2006)
My major research topic at BT was focused on the application of multi-agent systems technology (one of the most popular cooperating supporting tools) to mobile business processes. I was involved in 5th Framework European Project LEAP from 2001 to 2002. LEAP Project was completed by the effort of LEAP consortium that consisted of European Telecom Companies and Universities, and it produced the first FIPA compliant multi-agent platform for mobile devices. I contributed to the project by developing agent-based reusable service components for mobile applications in three major application domains such as travel management, knowledge management and distributed teamwork coordination. The developed service components were successfully re-used in the implementation of mobile applications to support the mobile business processes in BT and ADAC (a European car-breakdown service company). I was also involved in the mPower project which aimed to develop a component based application framework.
· Coordination of Inter Organizational Workflows (2003 – present)
Supply Chain Management has been one of the most popular research topics. We are interested in increasing the flexibility of inter-organizational workflows. FCM is a flexible conversation model that is used to specify a new transaction protocol (or conversation policy) in a dynamic supply chain. We also developed ADINS as an agent-based dynamic information network for supply chain management. In ADINS, agent systems adapt to changes in a supply chain configuration by selecting efficient suppliers with an iterative relaxation contract net protocol and the data envelopment analysis (DEA) technique to optimize the efficiency of the whole supply chain not individual participants.
The assignment of mobile tasks to a group of mobile workers is considered as one of the most challenging task due to the large number of constraints and the uncertainty posed in the constraints in the mobile working environment. The traditional and currently most popular so called ‘Big Brother’ approach is not appropriate to make optimal schedules based on the uncertain constraints. As a result, market based approach wherein workers make their schedules based on their local knowledge and handle exceptions by collaborating with colleagues receives a big interest among the researchers recently. However, the market based approach is based on the competition among the workers and it is not easy to be adopted in real organizations due to political issues such as the resistance from the Unions. This paper proposes a novel design of market based task scheduling system that considers social security which guarantees task schedules for the minimum level of income for all the participants. The design will be verified via a simulation that will identify the core variables and their relationships in the market system based on statistical models.
Many distributed scheduling mechanisms have been proposed to overcome the limitations of centralised scheduling mechanisms. Despite of the advents of many different types of scheduling mechanisms, the literature does not clearly reveal the pros and cons of each mechanism. This research aims to identify the conditions in which each mechanism shows better performance over others. So far, the comparisons of the mechanisms have mainly been based on mathematical models that have limitations on the identification of precise dynamics among relevant variables. Simulation based analysis can provide multi-dimensional analysis for the identification of the conditions. Furthermore, this research considers uncertainty such as the occurrence of exceptions in mobile working environment via the simulation.
The ad-hoc cross organizational workflow in a virtual organization is classified from others in that its workflow instances are sometimes executed based on different workflow schema. As a result, the exception handling of the ad-hoc workflows becomes more complicated than others as the designer of the workflows cannot predefine the exception handlers for all the possible exceptions for different instances. This paper proposes an agent-based exception handling approach wherein the intelligence of the agent can plan exception handling procedures for different workflow instances based on the basic principles given by a workflow designer. The proposed approach is applied to an eMarketplace example to show its validity, and the advantages and disadvantages of the approach are discussed by comparing existing approaches.
I have selected three papers among my refereed journal papers. The online files of the papers are available in Management Science, IEEE Pervasive Computing, Expert Systems with Applications, and BT Technology Journal.
· Realising Teamwork in the Field: Agent based Approach (Co-authored with P. Mihailescu and J. Shepherdson), IEEE Pervasive Computing, 6 (2), 85 – 92, 2007.
Reason: The paper identifies the major issues to support mobile workforces and proposes an information system to resolve the issues.
File Name: A-MPOWER-IEEEPC2007.pdf
· Exploration and Exploitation in the presence of Network Externalities (Co-authored with J. Lee and J. Lee), Management Science (SSCI) 49 (4), 553 – 570, 2003.
Reason: The paper shows an example of my future research in simulation based analysis on the network externality in a group decision making process. My future work in this area will be an extension of the research framework employed in this paper by adopting multi-agent based simulation to investigate the various side effects due to the interactions among participating business entities.
File Name: B-Simulation-MS2003.pdf
· An Agent-based Dynamic Information Network for Supply Chain Management (Co-authored with H. Ahn), BT Technology Journal (SCI) 22 (2), 18 – 27, 2004.
File Name: C-ADINS-BTTJ2004.pdf
· A Flexible Agent System for Change Adaptation in Supply Chains, Expert Systems with Applications (Co-authored with H. Ahn and S. J. Park), 25 (4), 603-618, November 2003.
File Name: D-FCM-ESWA2003.pdf
Reason: The two papers show how multi-agent systems technology can be used to increase the flexibility in supply chain management. The papers are challenging a structural flexibility issue of a supply chain.