Abstract
Predictive process monitoring techniques aim to forecast outcomes of running business process instances. These techniques are based on using predictive models built from past observed behavior, i.e., in an offline setting. However, business process behavior usually changes over time and predictive models are therefore at risk of becoming obsolete. Because of this, the definition of systems that build predictive models through an online setting has recently gained attention. Nevertheless, the scalability of this kind of setting within a context where the amount of data available is experiencing rapid growth is an outstanding issue. To solve this problem, this paper aims to define a framework for event sequence prediction capable of taking advantage of modern distributed processing platforms. An implementation over this framework based on Apache Flink is presented and it is tested upon two different case studies to prove its validity and its capacity to scale.
Highlights
Event data is everywhere, from isolated devices to entire organizations, and its extraordinary growth over the last decades will keep an exponential trend [1]
Every event within an event log refers to a case and a particular activity executed for that case
The sequence of events associated with a particular case can be seen as one execution of the business process managed by the event log
Summary
From isolated devices to entire organizations, and its extraordinary growth over the last decades will keep an exponential trend [1]. One of the main limitations of the techniques framed within these categories is that they are usually used in offline settings They feed from the historical information offered by event logs to provide knowledge about processes without considering running process instances. Rico et al.: BP Event Prediction through Scalable Online Learning logs that can be further used to predict the behavior and performance of running process instances [7] This has a crucial role in supporting organizations on proactive management tasks to improve process performance and detect risks [8]. The knowledge of such outcomes can be especially useful to support business processes and can serve as the first step to subsequent checks on process monitoring For this reason, through this paper, sequence prediction is considered as the main object of prediction.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
