Recently, enterprise information systems accumulate business events in log files. Such event logs enable analysts to keep the track of conducted tasks and context in business process execution. Process mining techniques are widely used to understand and analyze the event logs, considering their underlying business process. This research introduces a technique for bottleneck identification in business process using process mining and process capacity theory. Trace clustering is also applied to divide process instances in the event log into several process models which contain similar behaviors. The proposed technique is illustrated through experiments using real-world event logs which were collected from repair service execution in a telecom company.
목차
Abstract 1. Introduction 2. Framework 2.1. Trace clustering 2.2. Process model discovery 2.3. Process bottleneck 3. Bottleneck Identification 3.1. Data description 3.2. Process capacity 4. Experiment Results 5. Conclusion References
키워드
Process miningBottleneck identificationTrace clustering
저자
Gwangjin Heo [ Department of Industrial and Management Systems Engineering, Kyung Hee University ]
Jinsung Lee [ Department of Industrial and Management Systems Engineering, Kyung Hee University ]
Jae-Yoon Jung [ Department of Industrial and Management Systems Engineering, Kyung Hee University ]
교신저자
한국EA학회는 전사적 관점의 아키텍처 개념 및 원칙을 국내 민간기업 및 정부기관에 적용 확산시키고, EA 및 관련 분야의 연구, 전문인력의 양성 및 정책적 건의 등을 통해 기업 및 정부기관의 경쟁력 및 생산성을 향상시키고, 우리나라 지식 기반 산업 등의 고도화를 도모하는 것을 목적으로 합니다.