Predictive Maintenance vs. Preventative Maintenance

Technological advancements are changing the way the services industry operates. With the emergence of the Internet of Things, machine learning and artificial intelligence, technology has introduced new, more efficient ways for service businesses to maintain equipment and assets and improve their overall operating model.

One of the greatest technological advances of the last decade has been the Internet of Things (IoT), which is allowing businesses to access and analyze data that was previously inaccessible. With this new insight service organizations can drastically improve business processes and generate new business opportunities.

One of the primary benefits the IoT and related technologies offer is helping manufacturing organizations improve machine maintenance. According to McKinsey & Company, “The factory setting is one of the largest sources of value from the adoption of the Internet of Things, potentially generating an economic impact of $1.2 trillion to $3.7 trillion per year.”

Serving these increasingly IoT-connected, highly productive organizations has in turn created an entirely new service market – Maintenance-as-a-Service. Maintenance-as-a-Service (MaaS) allows service companies to provide a secondary service layer, creating a new revenue channel in which maintenance becomes its own service offering.

But before we get into MaaS and its impact on business, let’s first understand its foundation in the two core types of maintenance that exist today: predictive maintenance vs. preventative maintenance. Or, for a quick and easy way to distinguish the two concepts, reactive maintenance vs. proactive maintenance.  

Predictive Maintenance vs. Preventative Maintenance: What’s the Difference?

Predictive Maintenance: A technique that utilizes sensors to help determine the condition of any part or piece of equipment, predict breakdowns and pinpoint the most cost effective time for maintenance without losing any throughput. That is, it tries to predict failure before it occurs using continuous observations and health-checks.

Preventative Maintenance: A technique of mitigating equipment risk by performing regularly scheduled maintenance. The schedule is predetermined by forecasting the overall use and normal wear and tear within a given time period. Or to use an analogy, it’s the same idea as taking your car in for regular oil changes to preempt a total breakdown; also referred to as Planned Maintenance.

How Predictive and Preventative Maintenance Work

Defining and theorizing how these techniques works is great, but let’s look at two real-world examples and see how these techniques play themselves out. We will look at the following:

  1. Approach
  2. Cost (Customer and Business)
  3. Pros and Cons

Predictive Maintenance:

Let’s break it down via an automotive example we can all relate to – any individual with a car should have regular maintenance performed on their car to avoid erosion and increase longevity.

The Predictive Maintenance Approach:

Predictive maintenance argues that, instead of taking the car into the shop at a preconfigured time, we should be consistently looking at these parts at regular intervals to better understand how parts are eroding.

Now, most individuals neither have the knowledge nor the ability to take these readings; and taking the car in regularly to get these readings defeats the purpose of predictive maintenance, as it focuses on the importance of ongoing monitoring instead of regular, set checks.

One way to combat this would be if the major parts of a car, let’s say tires, had a device that was configured only to check tire pressure. This device would then do the following:

  1. Continually check the tire pressure for each tire.
  2. Gather and compare that pressure to the standard thresholds.
  3. Flag or not-flag potential areas for maintenance.

The Cost:

Based on the above tire readings over time, if we’re a car manufacturer or service shop, we can:  

  1. Better predict when a car would require a tire change.
  2. Inform the customer ahead of time and book an appointment in advance.
  3. Manage tire inventory more effectively.
  4. Staff the shop accordingly for upswings and downswings in tire change work.

The above allows me to not only reduce my inventory and staffing costs, but as a manufacturer I can better understand why certain tire brands or cars erode faster or slower, which gives me an extra advantage for catering to customers with specialized needs.

As the owner of the car, some of the benefits to me are:

  1. Maintenance is planned based on concrete data.
  2. Visibility on what needs changing when.
  3. Better planning and scheduling around maintenance.
  4. Customized service.

This reduces the customer’s cost by not having to conduct maintenance when it’s not required, and also allows for customized maintenance instead of having to purchase a defined maintenance package. 

Pros and Cons of Predictive Maintenance:



Cost effective



Higher initial investment

Reduces downtime or allows for better planning


Increases asset lifespan


Cuts routine maintenance costs



Improves quality of service delivery



New revenue streams for equipment manufacturers and service companies


Preventative Maintenance:

Sticking with the automotive industry example, we assume that a Toyota Prius’ brakes will last ~200,000 miles, or 12 years.

The Preventative Maintenance Approach:

When following a preventative maintenance model, brake maintenance for the above Toyota Prius’ brakes would be preconfigured to happen either after 200,000 miles, or 12 years,.

However, preventative maintenance fails to take into consideration that each Prius is driven in uniquely different circumstances. For instance, let’s consider differences in driving styles. An individual with an aggressive style of driving is constantly thumping the gas and then the brake; the constant thumping causes a higher stress on the brakes, thus causing them to erode or fail sooner.

In comparison, a passive driver who decelerates by lifting his foot off the gas instead (due in part to paying closer attention to traffic ahead and keeping a greater distances from other cars) causes the brakes to be used less frequently and relies more on gradual stoppage and momentum.

While we aren’t faulting either driver’s abilities, we can conclude that in the latter case, brakes might possibly last longer; while in the aggressive driver’s case, the brake pads may need to be looked at much sooner.

The Cost:

The main issue with preventative maintenance is that it looks to group together all similar parts or components of a machine but ignores how those parts or components are actually being used. That leads to the following costs for a car manufacturer or service shop:

  1. Too much or too little inventory.
  2. Staffing based on historical trends.
  3. Fluctuation in customer engagement.
  4. Poor service.

From a customer standpoint:

  1. Unnecessary maintenance (too early).
  2. Being unprepared; waiting too long, which could be very dangerous.
  3. Sporadic maintenance that doesn’t allow for effective planning.
  4. Non-customized service; more maintenance is performed than is necessary.
  5. Potentially lower quality of service.

Pros & Cons of Preventative Maintenance:



Increased uptime

More maintenance

Lower long-term repair costs

Higher inventory cost up front

Lesser likelihood of major failure

More volatility in staffing

Maintenance Trends

As we move into a more connected world with higher availability to data and a constant drop in costs per part but an increasing cost in labor service organizations are looking for any competitive advantage they can get and Predictive Maintenance is one of them. For service organizations to exist in the future world of interconnected technology, the ones that adapt now and are more geared to a MaaS model will allow them to not only be disruptive but also create a new market space to live in. This is the future of service, and the only way to get there and compete is to invest in technologies that get them closer to a Predictive model than a preventative model.

In a research paper by Dr. Mohan Kumar Pradhan and Mr. Jyotiprakash Bhol, the two point to the following time spent on maintenance:

  1. 39% of time spent on maintenance is for unforeseen repairs.
  2. 20% is for preventative maintenance.
  3. 37% for planned repairs
  4. Use of computers to control spare parts increased from 10% – 50% in recent years.
  5. Computer control preventative maintenance increased from 9%-60% in the same timeframe.

The above statistics illustrate how the world is changing around us, and where companies need to adapt to gain a competitive edge in a changing industrial market. The rise of connected parts/machines is enabling robust data gathering which is altering the current landscape and creating a new market: Maintenance-as-a-Service. 

Maintenance as a Service

As a result of the IoT and the possibilities the IoT has created, allowing machines and parts to connect, it’s become easier than ever before to collect large volumes of maintenance data. This data collection and analysis has led to the new, disruptive maintenance market – MaaS – that enables companies to:

  1. Predict the lifetime of a machine or part.
  2. Provide more curated maintenance (time and type).
  3. Create a deeper understanding of machines and parts based on how individual parts perform.
  4. Find outliers and ways to streamline bottlenecks.

This new paradigm is:

  • Allowing companies to closely monitor their machinery and parts from anywhere, at any time.
  • Creating smarter workforces, and also smarter factories and plants.
  • Creating a new avenue for manufacturers to ensure constant engagement with their clients by offering more curated service based on the data they collect and analyze.

Maintenance-as-a-Service (MaaS) allows both the customer, the new machine owner, and the manufacturer, to better understand when, why and how parts erode. With this information they can determine the best time for maintenance by scheduling a repair or fix based on data, rather than a pre-defined timeframe or run-time. 

How Hitachi Solutions Can Help

As a leader in the industry, Hitachi Solutions aims to be one step ahead, helping to provide customers with the latest and most innovative technologies. In order to ensure service organizations are utilizing the latest advances in technological innovation we created the Hitachi Solutions IoT Predictive Services Hub. The IoT Predictive Services Hub was built on the Microsoft stack of technologies, giving customers the familiar, user friendly technology they’re used to while delivering a 360-degree view of equipment and its maintenance needs.

The IoT Predictive Services Hub works by taking any IoT-enabled device and registering it to the hub. Once in the hub, the device’s information is gathered and stored in the cloud. Next, rules and thresholds for the device are created, with each rule and threshold containing one or more actions. Examples of actions might be, to send an alert if a part isn’t functioning as intended, schedule a repair, order a part, schedule a maintenance, etc. Once the data starts to gather, it begins feeding it into the Advanced Machine Learning tool, which will look to find trends on why parts fail, look for early indicators of wear and tear and look for anomalies that we have yet to encounter, along with other learnings. The data is all stored in the cloud, making this a fully cloud-enabled solution.

The IoT Hub also offers dashboards and analytics to showcase the rich data that is being collected. So now matter where you are located you can stay up to speed on the heatlh of your equipment. For instance, you could be at a plant in Mexico, and know how your plant in Texas is performing. The IoT Hub is able to create summary reports to showcase up-time, number of parts nearing maintenance, failures today, days to repair, etc.  

The Hitachi Solutions IoT Predictive Services Hub offers clients multiple avenues through which to approach the changing maintenance market. This solution also enables clients to provide their customers a better quality of service, learn more about their own machinery, create smarter work orders, plan on a more effective scale, and also create a new revenue stream that wouldn’t have been possible without the advancements in technology we see today.

Hitachi Solutions will be showing the exciting features of the Hitachi Solutions IoT Predictive Service Hub at the upcoming virtual event, The Manufacturer of the Future: How IoT and Predictive Analytics Transform Operations for the Digital Age, on Tuesday, May 2nd, at 9:00am-12:00pm PDT. Don’t miss out – click here to register!

For questions on the Hitachi Solutions IoT Predictive Services Hub and how it can impact your business, please contact Hitachi Solutions today.