Keeping equipment running at optimum capacity over an extended period of time is a conundrum faced by almost every plant operator.
In an effort to achieve this feat, production facilities will monitor everything from vibration and temperature right through to a machine’s acoustic readings.
I believe that in the future we will see the emergence of prescriptive maintenance strategies
Siemens vice president Andreas Geiss
Processing equipment firm Hosokawa Micron provides maintenance solutions across industries including chemical, food & beverage and pharmaceuticals, to help its customers better understand where the faults lie in their systems.
For instance, it offers clients a knowledge-based predictive maintenance app designed to remotely monitor the major parameters of a plant.
The app can also trigger alarms if those parameters are outside of normal operating ranges.
“We can then add in to this solution online trending, condition monitoring, online particle size, online advice and automatic report generation for quality control records,” says Ian Crosley, managing director of Hosokawa Micron.
As machinery becomes more advanced and factories increasingly intertwine more elements of their production processes, the maintenance strategies implemented by plant operators will have to become just as advanced if they have any hope of keeping up.
“Technologies need to be adaptive, and also ultimately will need to be predictive, and have the capability to be connected to other systems such as material requirements planning and enterprise resource planning systems,” Crosley says.
Maintenance providers aren’t stopping there, either.
The Internet of Things (IoT), which consists of any ‘thing’ that can be assigned an IP address, has the potential to substantially boost production capacity, reduce maintenance, and centrally monitor and control crucial operations within process facilities, he says.
As companies move to integrate systems across their entire operations, internet-led production will also have to work seamlessly with their maintenance regimes.
“The Internet of Things ultimately requires the integration of the total factory environment and should include all aspects of the company’s activities. This gives the maximum flexibility in being able to ramp up and down production to meet the needs of the order book,” Crosley says.
“This means that predictive maintenance has to become more science-based, rather than triggered by alarms set at specific values. The monitoring software has to be able to cope with a wide range of process parameters.”
Ultimately, Crosley says that predictive maintenance plays a vital role within an IoT strategy to help advance production capability.
“The adoption of the IoT needs the systems to be dynamic, not static,” he says.
Ian Pledger, field services engineer at Schaeffler UK, says IoT-based maintenance strategies will also allow systems to self-diagnose and talk to one another.
“Through the Internet of Things, you can have an array of sensors that will work to accurately predict machine failure at various points across your production line,” Pledger says.
Schaeffler’s SmartCheck sensor is designed to generate data on a continuous basis, picking up trends and any machine defects, within an IoT strategy, he says.
According to Pledger, if an engineer uses sensor-generated information to fix a faulty machine, the maintenance ‘loop’ is effectively closed – an idea that he says is twofold. Firstly, an engineer must act on the information received.
“However, if you don’t act on the information that is generated then it is worthless,” Pledger says.
Secondly, closing the loop validates the condition monitoring techniques being used - comparing the defect diagnosed from the data generated by the system with the actual failed component, he says.
“The IoT will eventually enable loop closing to be performed fully automatically, as the information obtained from an intelligent device can inform an engineer as to the parts that are needed to replace those that are defective.”
Siemens vice president Andreas Geiss says the IoT will dramatically change the ways in which predictive maintenance strategies are executed.
“I believe that in the future we will see the emergence of prescriptive maintenance strategies,” he says.
According to Geiss, prescriptive maintenance offers a plant operator more intelligent and strategic ways to analyse data, which will ultimately deliver a far clearer answer to the question: ‘what exactly must I do to correct the fault’?
This is in contrast to predictive maintenance, which provides an answer on what is likely to happen within a plant, he adds.
“Prescriptive maintenance is the next step in integration and will be the next quantum leap in reaching a higher productivity yield.”
Geiss claims that the move towards prescriptive maintenance strategies will be helped by the uptake of the IoT.
“It will drive the cyber-physical systems and will massively benefit process plant operators throughout the world,” he says.
However, Geiss says a change in culture will be essential if IoT-led maintenance is to succeed.
“The landscape from an educational and a human skill standpoint will ultimately have to change,” Geiss suggests.
“It will be more about having appropriately-educated people who are not necessarily in-depth specialists in one specific area of a plant or an asset. They will have a more general overview and a wider understanding of what needs to be done with all the data and information that smart maintenance systems produce.”
Like Pledger, Geiss is clear that the more data a plant converts into actionable information, the more it will benefit in the long run.
Geiss suggests it is a case of pure statistics and mathematics.
“As the information grows, there will be a need for further systems to digest the glut of information. People will be challenged to digest this information and will be tasked with gathering only the best data to make effective changes and keep plants running at optimum capacity.
“The view we have today of the typical maintenance engineer will fundamentally change,” he says.
German engineering firm Bosch Rexroth provides a number of maintenance solutions including systems analysis, condition monitoring and machine service contracts.
In a recent survey conducted by the company, it found that over 70% of the maintenance undertaken within UK firms was either planned or reactive, while less than 30% was preventative or predictive.
To combat this, and to build on its existing remote condition monitoring solutions, Bosch Rexroth is launching an advanced system that can analyse a wide range of data sets and learn from the information gathered.
Bosch Rexroth UK service manager Richard Chamberlain says the system is being designed around server-based software and has the capacity to take millions of data samples and teach itself.
“The analysis the software can conduct is within the parameters of the actual machine itself,” Chamberlain says.
According to Chamberlain, the new system will be unique in the marketplace.
“Other condition monitoring systems just give an alarm. Ours will be ‘fully teached’ and will understand how the system is supposed to work,” he says.