Maintenance-as-a-Service: what is MaaS?

It’s no secret that predictive and prescriptive maintenance still have a prohibitive cost for small and medium-sized enterprises. But what if you could have them as a service or as a subscription, much like your current CMMS (or our IMMP)? Or buy maintenance services from your equipment suppliers? IoT opens a whole new world of possibilities, and that’s what Maintenance-as-a-Service (MaaS) is all about.

 

How does predictive maintenance work?

Predictive and prescriptive maintenance are the future. Predictive maintenance consists in collecting data and using algorithms to predict failures. It requires condition monitoring hardware, connectivity, and advanced analytics – all of which come with a high bill. 

 

Prescriptive maintenance takes things a step further since it “prescribes” concrete maintenance actions (such as ordering parts, scheduling a repair, and so on). Of course, this requires everything predictive maintenance does, plus industrial automation hardware. 

 

One way to curb these investments and solidify predictive maintenance’s spot in “the mainstream” is Maintenance-as-a-Service. IoT will enable cyber-physical systems, with interfaces for accessing digital data and predicting the likelihood of failure. 

 

What Maintenance-as-a-Service will look like

When we envision Maintenance-as-a-Service (MaaS), it’s mostly “predictive maintenance as a service”. This would allow smaller businesses to try new maintenance strategies, and grant them access to cutting edge technologies without sinking themselves in debt and risking their sustainability. 

 

The most straightforward model for predictive maintenance-as-a-service is probably software mixed with machine learning. First, the software receives periodic or continuous equipment monitoring data. Then, based on historical data, and artificial intelligence, it does all the data science and computing work on its own. Accuracy is likely to improve over time.

 

Unless… we can bring manufacturers on board. Equipment manufacturers have the most information about assets. Their knowledge is pivotal to monitor assets and build accurate data models to predict the probability of failure. This brings us to other potential models of predictive maintenance-as-a-service.

 

Potential Predictive Maintenance-as-a-Service Models

 

Equipment health monitoring and maintenance recommendations as a service 

In the future, manufacturers might provide online tools (cloud-based tools, perhaps subscription-based) to monitor the status of their equipment. These tools allow companies to obtain predictive maintenance recommendations and adjust their plans accordingly. 

 

For manufacturers, this is an opportunity to upsell. Apart from selling equipment, they would offer maintenance services as well. For companies, this option is likely to provide detailed insights. Software based solely on historical data will only be able to produce higher-level maintenance recommendations (“schedule inspection” or automatic work orders for “calibration/ lubrication/etc).

 

Integrating equipment data with manufacturer data through maintenance platforms

If manufacturers don’t provide online tools, or if they can’t provide recommendations to each client, third parties may provide an answer. As an extension of current IMMP and CMMS software, equipment monitoring data can be integrated with the manufacturer’s data sets and recommendations.

 

Naturally, this requires collaboration between manufacturers and software providers. However, with the current emphasis on the right to repair and the urge for more transparency – car manufacturers already have to release maintenance plans, for example – it’s a feasible solution. 

 

Third parties will likely be able to customise their interfaces, and perhaps even algorithms, to each client. Therefore, this will likely be a more customer-driven option than the manufacturer’s tools.

 

Equipment as “a utility” and leasing options

The “sharing economy” has not yet made its splash in the maintenance world. But there’s still time. Instead of paying for predictive tools upfront, companies may pay them per hour or per uptime. Monitoring hardware becomes a utility or a subscription service.

 

In fact, equipment itself may be paid in monthly instalments (in a leasing arrangement). The manufacturer is responsible for its maintenance, and companies pay fees according to their usage, which is monitored through smart meters. This might be a solution to acquire 3D printing equipment, for example.

 

Sharing data with suppliers

More than “maintenance-as-a-service”, this would be “inventory management as a service”. You may share equipment data with suppliers to automate orders and avoid waiting times. If your suppliers can predict when you’ll need what, it’s much easier to manage inventory just in time.

 

Bonus: “Warranty-as-a-Service”

We’re all too familiar with the endless blaming game. When a machine fails before it was expected to, manufacturers blame the operators. The operators blame the manufacturers and demand their warranty rights. Since we’ll be able to monitor equipment closely, warranties may change.

 

Instead of a time-based warranty (one or two years), we might see usage or output warranties, for example. These “predictive warranties” again put the onus on manufacturers and allow companies to prove they’ve been complying with all the recommendations. 

 

Why isn’t Maintenance-as-a-Service happening right now?

Maintenance-as-a-Service is becoming more of a reality as we speak. Every day, we try to infuse our maintenance platform with artificial intelligence and provide smart suggestions. However, there are still a few obstacles that prevent MaaS from becoming as common as CMMS, for example.

 

One of them is the implementation of IoT and increased connectivity, which also requires 5G coverage. Although most companies are aware of the importance of predictive maintenance, many are faced with ageing equipment. It needs to be either retrofitted or replaced. 

 

The other obstacle is data ownership and privacy policies. Asset data is a highly sensitive topic, and data policies between companies, suppliers and other partners would need to be ironclad. Perhaps some companies wouldn’t even be willing to share their data in the first place, which is what ultimately enables manufacturers and developers to build accurate algorithms. 

 

Lastly, there’s the issue of profit margins, transparency, and accountability. While all of these models seem beneficial for companies, for manufacturers it’s a completely new way to operate. The cost of manufacturing certain machines might be too high to have a decent margin on a subscription-based service. Or perhaps they are not ready yet to comply with warranties and data ownership policies. 

 

However, one way or another, IoT will meet predictive maintenance one day. Meanwhile, find out how an Intelligent Maintenance Management Platform can automate maintenance and improve your daily operations.