Smart Cooling Equipment contributes to predictive maintenance

Increasing machine availability with edge and cloud computing

Siemens Elektronikwerk Amberg builds on AI and edge computing to predict and prevent their PCB cutting machine‘s costly standstills.

 

The cutting machine for separating printed circuit boards at the Elektronikwerk Amberg suffered sudden standstills every one to six months. Each downtime took up to two shifts to resolve. Early anomaly detection and timely maintenance was the only solution to preventing costly downtimes. Turning to AI, edge and cloud computing, EWA built an AI-driven predictive-maintenance edge application that successfully has eliminated unwanted cutting machine standstills.

MindSphere - together with Mendix and Industrial Edge - is part of Siemens Industrial IoT

  

Siemens Elektronikwerk Amberg

Manufacturing tomorrow's electronics today

Elektronikwerk Amberg (EWA) produces 17 million Simatic products a year at a manufacturing quality of 99.999 percent. The award-winning smart factory went operational in 1990. Since then, EWA has automated 75% of its value chain and increased its output by factor 13. Today, the digital plant exploits some 50 million process and product data sets in its ongoing effort to improve products, processes, and suppliers to reach Six Sigma quality.

Industrial Edge – Boosting machine availability and productivity levels

EWA relies on Industrial Edge to predict potential spindle failures. This integrated edge-computing platform by Siemens supplies the local processing power, high availability, and short-latency times that power EWA’s predictive maintenance application – without time-consuming and costly edge-cloud interaction.

Learn about Industrial Edge

Solution in a Nutshell

In an optimization model called closed-loop analytics, EWA analyzes process data at the shopfloor level to optimize an underlying process in real-time. EWA decided to build an early-warning system for cutter spindles based on the closed-loop analytics principle.

EWA started by transferring historical data from its cutting machine and cutting process to the cloud. Here, EWA leveraged the storage capacity and processing power of Siemens’ MindSphere platform to train an algorithm. EWA’s algorithm identified two critical parameters connected to unplanned spindle standstills – the spindle’s speed and power consumption. These parameters emerged as indicators of anomalies, such as the quantity of cutting dust accumulated in the cutting machine’s spindle bearings approaching critical levels. EWA placed the trained AI algorithm at the core of an edge application, which they in turn uploaded to an edge device close to the cutting machine.

The AI-driven application now collects relevant machine data, including spindle speed and drive power demand, and analyzes it to detect anomalies in spindles and alert operators to preventive maintenance needs. EWA continues to upload data and anomalies to the Mind­sphere platform to refine the algorithm and increase the precision of its predictions.

Project Scope

In its production line for S7-300, ET200, and other SIMATIC products, EWA has a PCB cutting machine deployed that fashions printed circuit boards in various form factors. The project’s primary objective was to build an early warning system at the edge-level that would help prevent the cutting machine’s frequent and costly unplanned standstills. The edge application was not only to alert operators to preventive maintenance needs. It had to send alerts early enough to accommodate night shifts in scheduling maintenance with the cutting machine manufacturer.

Moreover, the predictive-maintenance solution had to be cost-effective, secure, scalable, and open to further optimization.

Outcomes

MindSphere

2 days of warning

AI predicts the need for spindle maintenance 12 to 36 hours ahead of potential standstill

MindSphere

0 unplanned spindle standstills 

Since EWA launched its AI-driven prediction of potential spindle failures, incidents are down to zero

MindSphere

120,000€ of cost saved per year 

By eliminating unplanned downtime for 18 spindles, EWA‘s cost of availability losses is down by 120,000 Euro per year

Offers to replicate this use case

  Cloud computing:

MindAccess:

  

MindAccess:

  

MindAccess:

  

MindAccess IoT
Value Plan

Visual Analyzer

MindSphere

Predictive Learning

MindSphere

SIMATIC Performance Insight

  

  

  Edge computing:

MindAccess:

  

MindAccess:

  

MindAccess:

  

MindSphere

Industrial Edge Device IOT2050

MindAccess:

  

MindAccess:

  

MindAccess:

  

MindSphere

Industrial Edge Cloud Connector Edge App

MindAccess:

  

MindAccess:

  

MindAccess:

  

MindSphere

Flow Creator Edge App

MindAccess:

  

MindAccess:

  

MindAccess:

  

MindSphere

AI Runtime Edge App

Connect with an IIoT expert

Business case, scope, goals: There is no one-size-fits-all IIoT strategy. Explore how you can leverage the power of IIoT to drive the value of your operations and business model.