Predictive maintenance is without doubt one of the major trends in enterprise maintenance. With the advent of Industry4.0 and the expanded digitization of industrial processes, plant operators are now able to collect and analyze vast amounts of historic data about their equipment. This enables them to predict and anticipate when a failure is about to occur, which reveals unprecedented opportunities for cost savings and improved quality management. Enterprises can capitalize on these opportunities via predictive maintenance programs offered by equipment vendors and maintenance solution providers.
Predictive maintenance technologies such as automatic data collection and data analytics are maturing at a fast pace. However, the wider deployment and use of predictive maintenance requires the identification and validation of viable business models, which leads to tangible benefits for all stakeholders.
In other words, it is important to identify proper ways of selling and operating predictive maintenance solutions, in ways that benefit solution providers, plant operators and Original Equipment Manufacturers (OEMs).
Before we can explore the business of predictive maintenance, it’s important to understand the key stakeholders in the ecosystem of this revolutionary form of industrial maintenance, what roles they play, and what matters to them:
These stakeholders can leverage the benefits of predictive maintenance in various ways using different business development strategies. For example, solution integrators opt usually for a customer-driven approach to meet plant operators’ needs. On the other hand, machine vendors are also looking at utility-driven approaches that change the scope of their sales from selling equipment to selling maintenance services (i.e. Maintenance-as-a-Service).
Likewise, the various stakeholders set different goals for their predictive maintenance projects. For example, plant operators and supply chain partners target optimized productivity based on OEE improvements, while solution integrators emphasize customer satisfaction. At the same time, OEMs are also interested in building long-term relationships with their business partners that purchase and use their products.
Based on the above-listed considerations, a variety of novel business models are possible, which are in most cases changing the common ways in which maintenance services are offered. Most of these business models promote a Maintenance-as-a-Service (MaaS) paradigm, which is in-line with the Service-as-a-Product model that is currently trending in industry.
These models emphasize plant owners’ and solution providers’ interactions with OEMs after the sale of the equipment i.e. in the form of after sales services. Such interactions unveil opportunities for new recurring revenue streams as part of added-value services that are offered to customers.
These are five of the primary MaaS business models.
1. Equipment health monitoring and maintenance recommendations as a service
This business model offers plant owners online tools that enable them to monitor the status of their equipment and obtain predictive maintenance recommendations. These tools are offered as cloud services. Hence, the model involves sales of cloud subscription with every purchase of equipment. It is naturally offered as a complementary feature by OEMs to plant owners.
In this context, this model can be perceived as an up-selling opportunity for OEMs, while enabling plant owners to increase their productivity. It is also possible for solution integrators to provide custom health monitoring solutions to plant owners. However, they may be unable to provide detailed insights on status of the machines unless they collaborate with OEMs.
2. OEE risk as a service
An interesting variation of the maintenance recommendations as a service is the OEE risk as a service model. The latter provides higher level maintenance recommendations (e.g., when to schedule maintenance of a part), accounting for the overall OEE risk. It can be offered by the equipment vendor in a way similar to the recommendations service outlined above.
However, this model can be also built as a custom service that exploits data from multiple machines and equipment in the plant. In this case, it will be integrated by the solution provider and offered to the plant operator as a service.
In-line with the MaaS concept, OEMs could opt to sell uptime of their equipment instead of selling the product itself. This leads to a utility-based concept, where the deployer of a machine or part (such as an engine) doesn’t pay for the product, but for the time that the machine/part is used based on a per-hour-of-operation charge. In this model, the OEM undertakes all needed service and maintenance activities, which are covered based on the pay-per-use charges.
The uptime as a service model is quite different from the health monitoring service outlined above. Specifically, it is destined to replace the conventional equipment purchase model, rather than being a complementary value-added service.
In cases where the OEM’s machine can be shared across many end-users, the uptime-as-a-service model could be used to move industrial maintenance to the realm of the “sharing economy” like many other popular IT-based services like Uber and Air BnB. This could become possible, if plant owners become more interested in operating rather than owning the machine or equipment.
MaaS can also change the relationship between OEMs and plant operators, by redefining the traditional warranty. Warranty claims can create problems and result in finger pointing, as OEMs and plant operators tend to refuse responsibility for failures and malfunctions. But, with operational data collection and analysis, equipment vendors will be able to prove claims about the operation of the equipment using real data rather than assumptions.
Hence, both plant owners and OEMs will be able to prove whether the equipment has been operated correctly. This will greatly facilitate agreements on warranty claims, which helps reduce costs and build trusting relationships between stakeholders.
Consequently, OEMs can offer many different varieties of “warranty as a service”, such as warranty based on time (e.g., 2 years warranty) or based on usage (e.g., warranty for 5,000 hours of operation), or even based on combinations of time and usage based schemes. All these services can be based on the collection and analysis of historical data in order to identify and verify the cases where the equipment was not operated correctly.
The primary purpose of predictive maintenance is to minimize critical and unexpected failures. To serve this purpose, information can be shared with business partners to plan relevant supply chain management processes and ensure the availability of spare parts as needed.
This business model is implemented between the plant operators and its supply chain partners. Plant operators can automate orders and other related supply chain processes, which help eliminate the hidden costs of reactive maintenance such as inventory costs and higher prices for spare parts.
Such information can be shared by enhancing conventional supply chain management systems with additional automation (e.g., orders triggered by predictive maintenance insights).
As predictive maintenance technologies mature, the implementation of the above models becomes technically possible. Nevertheless, there are still several challenges to be confronted prior to their wider deployment and use, including:
It’s important to understand the benefits and technical implementation of predictive maintenance. To succeed with predictive maintenance, it’s equally relevant to reflect on viable business models that drive sustainable deployment and use in the context of your operations.
John Soldatos holds a Phd in Electrical & Computer Engineering. He is co-founder of the open source platform OpenIoT and has had a leading role in over 15 Internet-of-Things & BigData projects in manufacturing, logistics, smart energy, smart cities and healthcare. He has published more than 150 articles in international journals, books and conference proceedings, while he has authored numerous technical articles and blog posts in the areas of IoT, cloud computing and BigData. He has recently edited and co-authored the book “Building Blocks for IoT Analytics”.