Several airlines have already signed agreements with Airbus on predictive maintenance. This means that all the data from the aircrafts is directly send, stored and analyzed to predict when maintenance should be scheduled or when certain failures might occur. The system constantly monitors health and transmits faults, measurements and warning messages to ground control. This helps to better plan maintenance and prioritizes the steps that need to be taken. Experts estimate such an approach is likely to increase aircraft availability by up to 35 %. This all seems very nice, but what does this exactly mean for a company that wants to start offering predictive maintenance to its customers? How will this new way of working affect the workflow that is now in place? An overview of the advantages and challenges.
Before having a return on investment from predictive maintenance, you first must invest. Predictive maintenance depends on how good the system performs and how it will treat the incoming data. The biggest chunk of your investment will thus go to the ‘learning’ of the system. Basic data is of course available, but from that point onwards, the system must learn and will make mistakes. The good news is that it becomes smarter every time a mistake has been corrected, meaning that you never have to correct the system twice. A prepackaged solution isn’t an option in most cases, because every product and sector has its own particularities.
On the other hand, once everything is up and running smoothly, you can start cutting costs and optimize your maintenance. The system will anticipate possible problems or breakdowns and optimize maintenance cycles. This means that you can proactively optimize a technician’s planning, availability and presence at your customers.
Because of the digital revolution, a lot of people fear for their jobs, as technology could easily take over. But does this mean that people will lose their job? Not necessarily. When you look at a customer service center, it is obvious that not every position will still be relevant. Because the system will predict eventual failures or breakdowns, a customer probably will have to contact the customer service department less frequently. As the system will give an accurate overview of the problem, customer service technicians don’t have to analyze the problem anymore, as this is already partially done for them.
Does this mean that they have become obsolete? Not at all. They will be used to create new processes, that need a technical point of view. They will rather create real added value and optimize the processes itself than try to find out what is the customer’s problem.
Let us first have a look on how maintenance and repair issues are tackled today. First of all, the customer notices a problem with his product. This means he will have to call the customer service department and explain what he is experiencing. Depending on the problem, the call agent will create a ticket and send it to the right technical department. They will further analyze the problem and schedule a technician to visit the customer taking into account the Service Level Agreements. As you can see, this is quite a long process and in meanwhile the customer’s product or installation cannot be used. Since the problem intake relies on what the customer has told us and how the call agent interpreted and noted down this information there’s still a chance that the technician arrives with the incorrect spare parts or tools needed for the job because the problem was somehow misunderstood or information was lost along the way.
The predictive maintenance principle changes the way companies will interact with their customers. This means that the system will work the other way around. It is not the customer anymore who will call the company, but the company that will contact the customer to schedule a repair or maintenance based on predictions made by the system. The customer does not need to spend time contacting customer service and by predicting the maintenance or repairs, the product is less prone to breakdowns.
We can’t talk about predictive maintenance without talking about the Internet of Things (IoT). The IoT as such is not a revolution in this context; all the products and installation nowadays already contain ways to measure certain values in the product or installation that give indications towards possible failure, nothing new here. However, before the IoT a person needed to physically go to this product or installation to actually perform the reading of these measurements; a classical “pull” of information. What the IoT does is making sure that these measurements are automatically send, over the Internet, to a central system that stores and analyzes these measurements. So, IoT makes sure that the information is automatically “pushed” instead of a classical “pull” of information. Once the information of all measurements are stored this system can start doing predictions on these measurements. Like any other prediction system, the quality of predictions increases with the amount of data available, the more the better!
Predictive maintenance offers advantages for customers and companies, if you put in the effort of changing your business model and go from reactivity to proactivity. A well-oiled and fast service means happy customers. And you cannot only use predictive maintenance for aircrafts, as stated in the introduction, but basically for any installation your company installed at your customers. Maybe the start can be rough, but in the end, it will certainly pay off.
Customer Engagement Expert