The Cost of Maintenance
Machine maintenance plays a key role in ensuring smooth factory and equipment operation. Maintenance costs are a major part of the total operating costs of all manufacturing or production plants, eroding productivity and profits. On an average, inefficient factories can spend anywhere from 15% to 35% on machine maintenance. With increased labour cost, manufacturers cannot afford to be lax as far as machine maintenance is concerned. To make things worse, undue machine downtime results in manufacturers losing their competitive edge, something that is simply unaffordable in these COVID 19 pandemic times. Another important factor is human safety. In sectors like rail, road transport, and aviation, negligence in machine maintenance can lead to loss of life and injury, which is irreplaceable.
We are living in the age of Artificial Intelligence (AI), Machine Learning (ML) and Internet of Things (IoT), which is already transforming the business world. Industrial Internet of Things (IIoT) is a subset of the IoT and explores the industrial application of IoT. As people started connecting Things to the Internet, it was realized that this connectivity not only provides remote control over things, but also allows collecting and analyzing data from devices on the fly. Predictive maintenance, a subset of machine maintenance, leverages the Big Data generated by sensors connected to machines in fine tuning performance of industrial assets. In countries like India and China, which are manufacturing hubs because of their cost effectiveness, predictive maintenance can make a significant difference in production. However, before we delve further into predictive maintenance and tools that facilitate it, let us briefly understand more about machine maintenance in general.
Types of Machine Maintenance
As mentioned earlier, machine maintenance plays a vital role in increasing manufacturing efficiency and productivity. Machine maintenance can be sub divided into three categories – reactive (a.k.a. run-to-failure), preventive and predictive.
Reactive Maintenance – Reactive maintenance follows the age old adage – ‘if it ain’t broke, don’t fix it’. This kind of maintenance has been the traditional way of dealing with maintenance and sounds plausible on the face of it. A plant using run-to-failure management does not spend any money on maintenance until a machine or system fails to operate. This way of maintenance was acceptable till a decade back, when competition was not so intense, and markets were isolated. With the entry of players like India and China, companies cannot afford reactive maintenance now. The development of IIoT based instrumentation that can be used to monitor the operating condition of plant equipment, machinery, and systems, has provided the means to manage the maintenance operation efficiently. This instrumentation has provided the means to reduce or eliminate unnecessary repairs, prevent catastrophic machine failures, and reduce the negative impact of the maintenance operation on the profitability of manufacturing and production plants.
Preventive Maintenance – quite simply, preventive maintenance implies routine maintenance tasks that are based on elapsed time or hours of operation. Take the example of your car. Ideally, you should get it serviced after it has run certain number of kilometers, irrespective of whether it is giving you trouble or not. This increases the life of your car in the long run. Comprehensive preventive maintenance programs schedule repairs, lubrication, adjustments, and machine rebuilds for all critical plant machinery.
Predictive Maintenance - Predictive maintenance is the systematic application of condition based early warnings (like change in machine vibration pattern, increase in temperature, appearance of cracks, increase in machine noise or change in pitch, etc.) in the maintenance of the production assets under consideration. A system that provides an early sign of failure helps in mitigating machine failure. Including predictive maintenance in a comprehensive maintenance management program optimizes the availability of process machinery and greatly reduces the cost of maintenance. It also improves the product quality, productivity, and profitability of manufacturing and production plants. Unheard of earlier, predictive maintenance is getting traction thanks to advances in IoT and IIoT technologies, sensors and IIoT platforms like PTC ThingWorx that facilitate communication.
Five nondestructive techniques are normally used for predictive maintenance management: vibration monitoring, process parameter monitoring, thermography, tribology, and visual inspection. Each technique has a unique data set that assists the maintenance manager in determining the actual need for maintenance. When IIoT is implemented on machines, sensors gather real time data, which is processed by a IIoT platform like PTC ThingWorx to obtain the actual operating condition of critical plant systems and based on this actual data schedules all maintenance activities on an as-needed basis.
Implementing Predictive Maintenance
Today's progressive manufacturers from India and other countries leverage IIoT in implementing predictive maintenance. Different types of sensors are installed into machines to capture data about the piece of equipment to enable evaluation of the asset’s efficiency. These sensors include infrared analysis sensors, vibration analysis sensors, motor circuit analyzers, thermal sensors, oil and lubricant sensors, and other such sensors that capture respective parameters that determine the machine's health. Using an appropriate platform (like PTC ThingWorx), the data from this sensors is transmitted to a central database in real time (typically cloud based). Predictive algorithms, that compare the current state of the machine / asset with the ideal state, are used to evaluate if there is any significant difference between the actual parameters. Through this exchange of information, maintenance managers can see all physical assets as a whole, allowing them to make sense of what’s happening in the machines and identify any areas that require attention.
Advantages of Predictive Maintenance
The predominant advantages of predictive maintenance include the following:
The downside is that sensors are still expensive, so predictive maintenance is practical only if you have a big setup. Of course, as awareness about the benefits of IIoT grows and the technology becomes more affordable, all industries will switch to predictive maintenance sooner or later in order to remain competitive.