Using Event Logs to Model Interarrival Times in Business Process Simulation
The construction of a business process simulation (BPS) model requires significant modeling efforts. This paper focuses on modeling the interarrival time (IAT) of entities, i.e. the time between the arrival of consecutive entities. Accurately modeling ent
- PDF / 795,080 Bytes
- 13 Pages / 439.37 x 666.14 pts Page_size
- 50 Downloads / 234 Views
1
2
Hasselt University, Agoralaan – Building D, 3590 Diepenbeek, Belgium {niels.martin,benoit.depaire,an.caris}@uhasselt.be Research Foundation Flanders (FWO), Egmontstraat 5, 1000 Brussels, Belgium
Abstract. The construction of a business process simulation (BPS) model requires significant modeling efforts. This paper focuses on modeling the inter‐ arrival time (IAT) of entities, i.e. the time between the arrival of consecutive entities. Accurately modeling entity arrival is crucial as it influences process performance metrics such as the average waiting time. In this respect, the analysis of event logs can be useful. Given the limited process mining support for this BPS modeling task, the contribution of this paper is twofold. Firstly, an IAT input model taxonomy for process mining is introduced, describing event log use depending on process and event log characteristics. Secondly, ARPRA is intro‐ duced and operationalized for gamma distributed IATs. This novel approach to mine an IAT input model is the first to explicitly integrate the notion of queues. ARPRA is shown to significantly outperform a benchmark approach which ignores queue formation. Keywords: Business process simulation · Process mining · Interarrival time modelling
1
Introduction
Business process simulation (BPS) refers to the imitation of business process behavior through the use of a simulation model. By mimicking the real system, simulation can identify the effects of operational changes prior to implementation and contribute to the analysis and improvement of business processes [7]. A BPS model is composed of several building blocks such as entities, activities and resources [6]. This work is related to entities, which are dynamic objects that flow through the system and on which activities are executed [2], e.g. passengers when modelling an airline’s check-in process. As for each BPS model building block, several modelling tasks are related to entities [6]. This paper focuses on the entity arrival rate, i.e. the pattern according to which entities arrive in the process. Accurately modelling entity arrival is crucial as it has a major influence on process performance metrics such as the average waiting time or the flow time, i.e. the total time spent in the system. To identify an interarrival time (IAT) input model, i.e. a parame‐ terized probability distribution [3] for the time between the arrival of consecutive enti‐ ties, inputs can be gathered by e.g. observing the process. However, as process © Springer International Publishing Switzerland 2016 M. Reichert and H.A. Reijers (Eds.): BPM Workshops 2015, LNBIP 256, pp. 255–267, 2016. DOI: 10.1007/978-3-319-42887-1_21
256
N. Martin et al.
observations are rather time-consuming, the presence of more readily available infor‐ mation sources should be investigated. In this respect, process execution information stored in event logs can be useful. Such files, originating from process-aware information systems (PAIS) such as CRM-systems, contain events associated to a case, e.g. the start of a pass
Data Loading...
