Event record-based evaluation of business scenarios in the logistics domain using process mining

Process mining is an upcoming field in data science using event logs. Current methods of evaluation of business scenarios in the logistics domain focus on an ex-post evaluation in set time frames, making it not applicable in reallife scenarios. This study proposes an event record-based model of evaluation of business scenarios in the logistics domain, and in addition, proposes how to select a better scenario if the current one is subpar. We do this by first executing a literature review to classify what different attributes can be present in a log, and which KPI’s could be used to quantify the success of a scenario, to finally propose a model that can be used for logistics use cases. With this information, a way to set qualifying criteria for a business scenario is described. The final model will be validated by a case study. This research proposes a method to link attributes to KPI’s, and the case study proves the proposed model suitable for the event record-based evaluation of business scenarios, taking a step towards filling the gap in research in the managerial point of view in process mining.