Blog

Optimised timing of maintenance increases and enhances the utilisation rate of equipment

The reliability of equipment forms the foundation of the successful operations of companies. Profitability correlates with operational reliability in all activities, such as travelling by plane or train, or transporting materials inside a mine or above ground, and in the effectiveness of industrial processes. Here, maintenance and its optimisation are of key importance.

Maintenance has traditionally been implemented based on an annual plan. The problem is that maintenance needs do not always follow the calendar. This may result in unnecessary stoppages of productive equipment. Alternatively, incorrectly timed maintenance measures can result in unpredicted downtime, which is expensive and has a negative impact on customer satisfaction. Therefore, maintenance means more than a simple oil change: it often concerns various extremely complex and expensive parts that wear in production, the replacement of which should be optimised.

Predictive maintenance that is based on sensor data can save millions of euros. IBM’s SPSS Modeler turns the data collected from sensors that monitor the processes and the equipment into a maintenance prediction model that supports operations. By predicting the wearing of the components, the maintenance model can be used to optimise the timing of maintenance measures. This reduces the risk of unpredicted stoppages and, on the other hand, unnecessary maintenance measures and related downtime.

Sensor data makes it possible to create a life cycle model for each part, which will keep track of parts that are likely to need maintenance in the near future and also to learn how unexpected stoppages of equipment can be prevented. Analyses enable the optimal operation of real-time alarms, reducing the workload caused by false alarms. Therefore, predictive and optimised maintenance can significantly reduce costs, enhance operational reliability and avoid unnecessary replacement of parts that are still functional.

For example, a large railway operator has installed thousands of sensors along a 20,000 km stretch of railway track to monitor the condition of carriage wheels and axles. With SPSS Modeler it is possible to utilise the information submitted by the sensors regarding critical parameters, such as the temperature, wearing, and faulty positions of the wheels, to predict and optimise maintenance and inspections. This has increased the operating hours of the rolling stock, and has also prevented unpredicted stoppages. The smart monitoring and maintenance system has increased the level of services and customer satisfaction among both business and leisure travellers.

When the aim is to create a smart maintenance model, SPSS Modeler is an effective solution for a large number of business areas. “The most simple and effective way of fixing a problem is to prevent the problem in advance.”