Process mining techniques have been developed in order to help process analysts to obtain better insights into the performance of business processes. A key dimension of such analysis is time, as it is closely connected with performance problems such as waiting times or bottlenecks. So far, process mining techniques have mostly focused on behavioral patterns such as directly-follows relationships, missing the opportunity to show time explicitly in generated process models. In this paper, we argue that representing time in process models is essentially a layout problem. We present a technique for aligning a process model with the time axis and exemplify it for directly-follows graphs. We use an illustrative evaluation using BPIC event logs to investigate the plausible benefits of a timeline-based layout.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Towards Timeline-Based Layout for Process Mining

  • Harleen Kaur,
  • Jan Mendling,
  • Timotheus Kampik,
  • Christoffer Rubensson

摘要

Process mining techniques have been developed in order to help process analysts to obtain better insights into the performance of business processes. A key dimension of such analysis is time, as it is closely connected with performance problems such as waiting times or bottlenecks. So far, process mining techniques have mostly focused on behavioral patterns such as directly-follows relationships, missing the opportunity to show time explicitly in generated process models. In this paper, we argue that representing time in process models is essentially a layout problem. We present a technique for aligning a process model with the time axis and exemplify it for directly-follows graphs. We use an illustrative evaluation using BPIC event logs to investigate the plausible benefits of a timeline-based layout.