Modeling time to failure using a temporal sequence of events

TL;DR

This paper introduces a new predictive model for estimating the time to failure of industrial machines using sensor event sequences, aiming to reduce downtime and maintenance costs.

Contribution

The paper presents a novel, efficient model that predicts machine failure time and identifies key sensors, improving over existing methods for real-time industrial machine monitoring.

Findings

The proposed model outperforms popular competitors in prediction accuracy.

It effectively identifies critical sensors associated with failures.

The approach can help prevent failures by controlling operational parameters.

Abstract

In recent years, the requirement for real-time understanding of machine behavior has become an important objective in industrial sectors to reduce the cost of unscheduled downtime and to maximize production with expected quality. The vast majority of high-end machines are equipped with a number of sensors that can record event logs over time. In this paper, we consider an injection molding (IM) machine that manufactures plastic bottles for soft drink. We have analyzed the machine log data with a sequence of three type of events, ``running with alert'', ``running without alert'', and ``failure''. Failure event leads to downtime of the machine and necessitates maintenance. The sensors are capable of capturing the corresponding operational conditions of the machine as well as the defined states of events. This paper presents a new model to predict a) time to failure of the IM machine and b) identification of important sensors in the system that may explain the events which in-turn leads to failure. The proposed method is more efficient than the popular competitor and can help reduce the downtime costs by controlling operational parameters in advance to prevent failures from occurring too soon.