It is possible to repair a factory machine after it has already broken down, but this is fraught with downtime, a decrease in production, or even an accident.
It is much easier to maintain the equipment before it breaks down, eliminating weak points – then you will not have to stop production and eliminate the consequences of the failure. For this, predictive maintenance of equipment is needed – we will tell you what it is and how it works.
What Is Predictive Hardware Maintenance?
Let’s take a look at one example. According to statistics, aeroplanes are one of the safest modes of transport. You will be in an unpleasant situation with a car on the way to the airport than to face technical problems during the flight. Predictive maintenance is one of the factors that make aircraft so reliable.
Aircraft engines are complex structures on which human lives depend. Their condition needs to be carefully monitored, which is why major aircraft engine manufacturers are building many different sensors into them.
During the flight, these sensors continuously write the operating parameters of the system into the internal memory. On the ground, flight data is sent to the manufacturer’s server, where it is analyzed. Then the airline-owner receives recommendations or requirements – depending on the severity of the possible problems – about replacing or repairing certain engine parts.
Aeroplanes require attention and care and any mechanical and electronic equipment and engineering structures: buildings, pipelines, cell towers, pumps, loaders, dams, machine tools, etc.
For large companies with a vast fleet of equipment, structures, many kilometres of pipelines and offices across the country, it isn’t easy to organize the maintenance of each piece of equipment – this requires tangible costs, organizing inspection schedules, and special people in the state. Even with all these efforts, significant breakdowns do occur and disrupt the ordinary course of work.
Predictive equipment maintenance is used to minimize the risk of sudden breakdowns and reduce the cost of maintaining large fleets of structures and equipment. The bottom line is that sometime before the system fails; you can find small signals that indicate problems.
Small voltage surges, deviations in the number of revolutions per minute, temperature out of permissible values, delays when starting the engine, incomprehensible noise or tapping, an increase in the liquid level, minimal changes in the angles of structures – by such signs, one can understand that a breakdown will happen soon.
Predictive Maintenance And Work Safety
Here are some examples of how to use predictive equipment maintenance in practice to prevent accidents and disasters:
- In St. Petersburg, there is a vast building – Lakhta Center, a height of 462 meters. To control the reliability of the structure, several thousand sensors are hidden in the foundation and walls of this building: pressure sensors, deformation sensors, deviations from angles, humidity and temperature. They transmit information to analytics servers, where the soil under the foundation is monitored and how the tower sways depending on temperature, humidity and wind speed. This allows you to monitor its condition, including during emergencies, such as strong winds.
- At the Smolensk nuclear power plant, engineers must bypass equipment and check its operation every day to prevent accidents or system failures. It used to take a full working day to check. To simplify the work of engineers, they were given special devices. They help determine the optimal bypass route, and the collected data is immediately transferred to the data centre for analysis. The analytics results are fed back to the engineers to decide if repairs are needed somewhere. This approach saved the company 45 million rubles a year.
- Shell is a large energy company that uses predictive analytics and services to identify and alert employees to security threats. This is how they manage to react to the problem before the accident happens.