skip to Main Content
While predictive maintenance is a ‘trendy’ topic in the industry currently, it’s important that companies do not deploy such system just to have it in their plant.

Predictive Maintenance for the Metalworking Industry

Predictive Maintenance for the Metalworking Industry

While predictive maintenance is a ‘trendy’ topic in the industry currently, it’s important that companies do not deploy such system just to have it in their plant. With clear goals and proper planning during implementation, companies should achieve better maintenance schedule planning, reduced downtime and better spare parts management. Article by David Chia, Beckhoff Automation Pte Ltd.

Preventive maintenance has been the industry’s default practice for maintenance for many years, where operators change out parts based on its estimated lifetime before the machine is expected to fail. Though preventive maintenance did help in reducing overall downtime of the machine, but it may not be completely optimised. Unnecessary change of parts may occur since the expected lifetime of a component is largely based on vendor’s recommendation and historical data, not taking into consideration other factors such as machine usage, which will also affect the lifetime of a component.

With such observation on the effectiveness of preventive maintenance, the industry then starts to look into predictive maintenance instead. As the term suggests, this approach predicts the machine failure and maintenance is carried out only when a failure is predicted to happen. A system will monitor the machine condition during its daily operation and alert the operators when an abnormal trend or signal is observed.

Along with the trend of Industry 4.0, predictive maintenance became a widely discussed topic among the various industries, including the metalworking industry. According to the Singapore Smart Industry Readiness Index (SSIRI), under the “Intelligence” Pillar of the “Technology” Building Block, one of the criteria states that a Smart Industry needs to be equipped with systems that are able to predict potential failures ahead of time and alert operators of such deviations, across the “Shop Floor, Enterprise and Facility” Dimensions. To achieve this, huge amount of data needs to be collected from the shop floor level and to be sent to the enterprise level for further analysis for failure prediction.

Therefore, predictive maintenance solution actually cuts across the shop floor and the enterprise level. At the shop floor level, companies need to look at what data are to be collected and how this can be done. For existing machines in the plant, it may means adding on new sensors to the machine to monitor parameters such as chamber temperature, vibration and pressure. Constant monitoring of these parameters will then generate sufficient data to determine a normal or safe machine operating condition, and able to detect out of bound signals and trigger warning to the operators.

These monitored trends are then sent to the enterprise level for data storage and analysis. With good amount of data over time, the system will be able to identify certain patterns of the machine operation, which can then be used to generate reports on more accurate maintenance schedule and identify areas of improvement. If we broaden the scope further, such data and reports can also help improve the spare parts stocking flow of the company, to determine what are the more common parts to stock, and the purchasing of spare parts can also be automated.

Although predictive maintenance has become a hot topic in the industry, there remains some hurdles to cross after a company has decided to embark on such implementation. Before jumping into conclusion hastily and decide on a predictive maintenance system, the company should always ask these questions. What is your current practice for machine maintenance? What are the areas that can be improved in your daily maintenance practice? How can predictive maintenance help you in achieving the improvement that you intend to see? Without clear goals being set at the start of the journey, companies will fall into the trap of just implementing a system without reaping the full benefits of it, and thus leads to waste of resources.

There are many predictive maintenance solutions available in the market. From standalone Condition Monitoring Systems (CMS) that consists of dedicated hardware and software, to monitoring features that are integrated together with the machine controllers, such as PLC, PAC or PC-based systems. While both are able to achieve data collection and analysis, standalone CMS are considered quick to deploy solutions since the software mainly involves configuration instead of programming. However, integrating predictive maintenance features onto a machine controller provides flexibility in terms of expanding the machine monitoring scope throughout different phases, and integration of more features besides analysis for predictive monitoring. Each of these solutions will have its pros and cons, ultimately it very much depends on the goal that the company has set and the return of investments (ROI).

While predictive maintenance is a “trendy” topic in the industry currently, it’s important that companies do not deploy such system just to have it in their plant. With clear goals and proper planning during implementation, companies should achieve better maintenance schedule planning, reduced downtime and better spare parts management.

 

CHECK OUT THESE OTHER ARTICLES

Heimatec: CyberCon4 Sensor System
Airbus Chooses China Airlines As MRO Supplier
Predictive Maintenance Solutions For Machine Tools Reduce Cost The Smart Way
EuroBLECH 2018: Hypertherm Launches Two Metal Cutting Innovations

WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!

FOLLOW US ON: LinkedIn, Facebook, Twitter

 

Smart Data in the Metalworking Industry
Putting Automation in the Hands of the People with Collaborative Robots
Back To Top