With the global situation weighing on supply chains and recession looming over the economy, more and more companies are looking for ways to get ready and weather the storm.
Agility may be one of these, for it enables fast results in dynamic environments and allows projects to pay for themselves quickly.
Another possibility, along with agility, we mentioned in our Tech Trends 2023 article, could be mixed teams and insourcing.
One we have not discussed yet, however, is predictive maintenance.
But what is it, and how can you take advantage of it in tough times?
A short definition
Predictive maintenance is a proactive approach to maintaining equipment that utilizes various technologies to predict when pieces are likely to fail or require maintenance.
By identifying potential issues before they occur, predictive maintenance helps companies avoid costly downtime, extend the life of their equipment, and reduce the inventory of spare parts. In today’s uncertain business environment, implementing a predictive maintenance strategy can be a key component of a company’s risk management and budgeting plan.
Concrete examples
There are several common technologies companies can use to approach predictive maintenance:
- Artificial Intelligence (AI) and the Internet of Things (IoT), as we explored in our project for ÖAMTC, can play a key role in predictive maintenance. In this case, the car´s software sends live data to the AI, which analyzes it and sends it to the server. There, the servers provide a consumable API that allows technicians to always know the status of the car, as well as to access logbooks for each vehicle. This way, it becomes possible to predict breakdowns and failures and intervene before they happen.
- Vibration analysis involves monitoring the vibrations of machinery to detect changes that may indicate a potential failure. Sensors attached to the machinery and connected to a monitoring system do this job nicely. The monitoring system analyzes the data collected by the sensors and generates alerts when it detects any changes that may indicate a potential issue.
- Non-destructive testing (NDT) bears prevention possibilities, enabling manufacturers to push the boundaries of their products. This enables them to know more accurately the lifespan of their parts, and what are the most common failures. NDT can also be enhanced with technologies such as Augmented Reality.
- Ultrasonic testing uses high-frequency sound waves to detect cracks or other issues within the equipment. Handheld ultrasonic testing equipment or fixed ultrasonic testing systems installed on the machinery are a proper setup. Ultrasonic testing detects issues in welds, bearings, and other components that may be difficult to inspect visually.
- Digital twins can allow, for example, B2B vehicle fleet operators to find faults by remotely accessing each vehicle. This saves precious resources in terms both of time and budget, allowing operators to always assess their vehicles´ status.
- Oil analysis involves analyzing the oil used in machinery to detect contaminants or other issues that could lead to equipment failure. This can be done using oil analysis kits or by sending a sample of the oil to a laboratory for testing. By regularly analyzing the oil used in equipment, companies can identify potential issues before they lead to equipment failure.
- Sensor technology combined with IoT also belongs in the predictive maintenance conversation. In the case of emergency vehicle manufacturer Rosenbauer, the latest generation of vehicles is equipped with up to 250 sensors. Together, they collect enough data to enable R&D work, improve daily operations, and – you guessed it – predictively maintain the vehicles.
Benefits of predictive maintenance
Predictive maintenance offers several benefits over legacy reactive maintenance, in which equipment is only repaired or replaced when it fails. One of them, for example, is that it helps companies avoid unexpected downtime. Additionally, maintenance work can be better scheduled and therefore reduce efforts and costs: When compared with corrective maintenance, predictive maintenance activities can be planned ahead allowing for better workforce planning and reducing unplanned work peaks that cause extra hours. When compared with preventive maintenance with relatively short periodic intervals, predictive maintenance can significantly increase the intervals and optimize the use of parts in terms of their actual lifespan.
Predictive maintenance can also help extend the lifespan of parts. If companies can identify potential issues before they occur, as we saw in the examples above, they will take steps to address them, proactively acting against the damage of parts. This can help reduce the need for expensive repairs or emergency replacement of parts.
Predictive maintenance can also help when supply chains are overloaded, as it allows companies to plan and order replacement parts in advance. This becomes paramount when dealing with hard-to-find or specialty parts, as it can take longer to source these items.
Extra benefit: Combating the rise in interest rates
News from the global finance world hasn´t been encouraging in the past months.
As interest rates rise, keeping a stock of parts can become more expensive. Some of the variables in place here are borrowing costs, storage and insurance, damage risk, and more.
When interest rates are low, companies usually find it more affordable to maintain a large inventory of spare parts. This of course is not the case anymore, considering the recent interest rate hikes by the ECB in Europe and the FED in the United States.
By implementing a predictive maintenance strategy, companies can reduce their need for a large inventory of spare parts, as they can instead rely on ordering parts as needed based on predicted maintenance findings. This can reduce the financial burden of maintaining a large inventory and free up resources for more critical business needs.
Taking your first steps toward Predictive Maintenance
If your company is interested in implementing predictive maintenance, there are several steps you can take to get started.
First, it is important to assess the current state of the maintenance processes and identify use cases where predictive maintenance could be beneficial.
This may involve conducting a thorough assessment of equipment and identifying which pieces of equipment are most critical to operations. This often happens with internal experts from production and maintenance departments that know the equipment by heart. As predictive maintenance is a data-driven approach, data scientists should be included as early as possible: On the one hand to prevent blind spots and potentially biased decisions by individuals. On the other, to conduct pilot projects as early as possible to test out different technologies and methods to determine which ones are most effective.
From our experience, agile mixed teams perform very well in this context, consisting of experts from production and maintenance as well as data scientists: The experts get a more objective data-driven view, learn new technologies and build trust in the solution. Data scientists can build on the experts’ knowledge to better target their analysis, discuss potential findings and evaluate first solutions.
As with any new initiative, it’s important to establish clear goals and metrics for success to ensure that the predictive maintenance strategy is effective and meets the needs of the business.
In doing so, keep balance in mind: Having a goal and KPIs is necessary, but an agile mindset is essential in these projects to make sure said objectives remain fluid and adaptable.
Staying ahead of the curve
In today’s uncertain business environment, implementing a predictive maintenance strategy can be a key component of a company’s risk management plan.
If your company is not already utilizing predictive maintenance, now is the time to get started if you want to position your company for success in the uncertain times ahead.
Do not wait until equipment fails or supply chains are overloaded – take the initiative and therefore control of your maintenance needs today.
