Ein Ansatz für ein Predictive-Monitoring-System zur Identifikation von Störungswirkungen in der Produktionslogistik mittels künstlichen neuronalen Netzen

Abstract

Unexpectedly occurring logistical and production related disruptions are part of the day-to-day operations of companies and have a negative impact on their process of service provision. The predictive identification of potential disturbances in the form of an early warning can help to extend the action period for countermeasures so as to counteract the actual disruptive effect in good time. In the age of increasingly large amounts of data about operational processes as well as providing information in real time, the use of predictive methods seems particularly promising. This article presents an approach to a predictive monitoring system (PMS) for identifying disruptive effects in production logistics. The core function of the system is based on artificial neural networks.