Maintenance Management and the Internet of Things

The last several decades have seen the rise of artificial intelligence, cell phones, the Internet, and other technologies that have dramatically changed the world. One of the “other” technologies, the Internet of Things (IoT), was introduced almost at the same time as the Internet and only a decade after the cell phone. While already a major industry in dollar terms, IoT still hasn’t had the public awareness and impact of the cell phone and the Internet. That is changing, as will be shown. In the first section of this document, we will look at the Internet of Things itself to define and describe it using the example of a dam and the body of water the dam controls.

We are also interested in Maintenance Management, which seems to be a natural fit for IoT, with devices containing sensors embedded in machines or other entities that provide readings that they send to computing devices in the Internet cloud. In the section on Maintenance Management and IoT, we go through the types of Maintenance Management and describe how each type is affected by IoT.

While successful, IoT has not progressed as fast as cell phones or the Internet. The following section discusses the roadblocks holding back IoT, in particular the precarious security of the “things” on the Internet of Things. From a moving vehicle to a cell phone to a body of water to a human body, practically any “thing” on the Internet of Things can be hacked with potentially deadly effects.

Finally, we talk about the future of the Internet of Things. While it might be fun to talk to engage in science fiction-type predictions, what emerges from IoT in the next few years could be life-changing to all of us, and we will concentrate on that.

Table of Contents

The Internet of things

The Internet of things is a complex subject, but we will keep things as simple as possible. To sort things out a bit, we will look at a dictionary definition of IoT and then break that definition down. From

The interconnection via the Internet of computing devices embedded in everyday objects, enabling them to send and receive data.

Note that the definition uses the word “objects” in place of the word “things. “Also, note that the word “machines” is not used instead of “things” or “objects. “Machines” are just a subset of “Things.” We can think of a thing as being anything of a physical nature, not just a machine which is what we usually think of, but also a building, a body of water, even a human being, and more.

To emphasize that the Internet of Things is not machine-specific, let’s look at the example of the body of water. The people who manage a lake might be interested in one or more of the features of the body of water – the number of fish, the level of algae, the pollutants in the water, the temperature of the water, the water level (the height of the water above sea level), etc. In addition, they may wish to take pictures to visually detect changes in the condition of the body of water.

To check these conditions, workers are stationed by the body of water to take readings or pictures of the water. Alternatively, if it is in a remote location, the workers might travel to the location when needed, get the readings, and then leave.

In addition to taking readings, the workers may be empowered to actually do something to the body of water. For example, let’s say the body of water is a reservoir created when a dam was built. If the water level is too high, the workers could release water and lower the water level.

All of this could be expensive. There is the cost of the travel and the living quarters for the workers for a remote site. In the days prior to IoT, these expenses might be tolerated, or perhaps the ongoing effort might be underfunded with poor results.

IoT might change this cost equation, so we will investigate how this works. The reservoir can be a thing on the Internet of Things, but it needs to be a “smart” thing. How is it made “smart”? Using the IoT definition above, it needs:

Computing devices embedded in everyday objects [the reservoir], enabling them to send and receive data

What might these computing devices be? At the base is a sensor. A sensor does pretty much what the name suggests – it senses something. In that way, a dog’s nose is a sensor – it senses odors as well as other things, such as heat. Unfortunately, as far as I know, there is no way to hook up a dog’s nose to the Internet to communicate what it is smelling to the rest of the world. This suggests one of the characteristics of a sensor useable for IoT – the system must be able to convert the data from the sensor into a form that can be transmitted over the Internet.

For our example, the computing device is a sensor for measuring the height of the water. At this point, the managers of the reservoir want the data from the sensor stored somewhere in a format that can be analyzed by a human or by a computer program that works without human intervention. Typically, this data is stored in a computer text or other type of file, a spreadsheet, or, in most cases, a database. Then, using the data in the database, if the computer program decides to release water from the reservoir, a message can be sent back electronically to a device in the dam to release the water.

Note that all of this can be done without access to the Internet. A local area network or wide area network can be used instead. This is important because, in practice, the operation of the various technologies may look very similar, and people may get confused as to what is being done using IoT and what is not. We may be overestimating the use of IoT.

But if the Internet of Things is used, here is a simple diagram of the flow of data:

Reservoir -> Sensor embedded in Reservoir (Water height sensing device) -> Router to Internet -> Internet -> Cloud -> Database within Cloud.

The word “Cloud” is a bit nebulous, but here is a typical definition from CloudFlare:

“The cloud” refers to servers that are accessed over the Internet, and the software and databases that run on those servers.

For our example, the most important thing about the cloud is that it contains a database where water level readings can be stored.

Is there a real-life example of a body of water that is a “thing” in the Internet of things? Here is an example of an initiative to digitally connect all of the earth’s water bodies; oceans, lakes, streams, and rivers.

I’ll note here that this is a really dumbed down (but hopefully understandable) example of how the Internet of Things works. If you are interested in a more complete understanding of the Internet of Things, here is one good reference: Internet of Things for Architects: Architecting IoT solutions by implementing sensors, communication infrastructure, edge computing, analytics, and security.

There is one more thing I would like to clarify about the Internet of Things. The word “smart” used in reference to a “thing” in IoT can be confusing. We like to think of humans as being smart because they have brains, but those brains don’t have anything to do with the Internet of Things. Instead, it is the implants (sensors) put into us that make humans the “things” the Internet of things. One example from ZDNet describes how the author has an implant in his chest to monitor an atrial fibrillation condition that has been corrected. The entire article is interesting, with the author telling how he thinks the individual will evolve from having physical keys, wallets, and cash to having implants that can replace all these things.

The use of the word “smart” in the expression “smartphone” may be even more confusing. Here the word “smart” refers to the computer in the phone, which has nothing to do with the Internet of Things in the same way that the human brain has nothing to do with the Internet of Things.

Can the smartphone be a thing on the Internet of things? Yes, it can, but only if it has a sensor embedded in it that relays information to the cloud – for example, the “find my phone” feature works using the Internet of things. There is a location sensor in the phone that is tracked by an entity in the “cloud” that can report to the user where their phone is.

Now that we have some understanding of the Internet of Things, we can show how IoT affects the various types of maintenance.

Maintenance Management and the Internet of Things

First, what is maintenance management?

The dictionary Merriam-Webster defines maintenance as

the act of keeping property or equipment in good condition by making repairs, correcting problems, etc.

and management as

the act or skill of controlling and making decisions about a business, department, sports team, etc.

Put together, Maintenance Management is the art and science of making decisions as to how to keep property or equipment in good condition. For the Internet of things, property or equipment is the “thing” in IoT.

In many cases, it turns out that maintenance management is a natural application for the use of IoT. After all, maintenance management is about physical things, the base requirement for being a thing on the Internet of things. It is also about measuring or inspecting the attributes of the thing, perhaps its temperature, vibration level, or in the example we used in the first part of the article, water level, and acting on those measurements if they indicate something is wrong.

Determining when and what maintenance to perform will differ depending on the equipment in question and the feature maintained. Looking at the maintenance management literature, there are as few as two or as many as nine more types of maintenance. Here we will concentrate on three commonly used types:

Reactive Maintenance

Reactive maintenance might be considered the absence of maintenance management. Things aren’t fixed until they are broken. This might be costly – changing the oil in a car is cheaper than replacing the engine. For this reason, Reactive Maintenance is often considered a bad way of doing things.

In some cases, it makes sense to deliberately let things fail rather than fix or replace them before failure. This maintenance management technique is called Run to Failure. Take, for example, the lightbulb. In most but maybe not all cases, it is realistic to let the lightbulb fail before replacement. The light bulb itself is low-cost and easily replaceable.

Both “reactive” maintenance and “run to failure” maintenance are poor candidates for the Internet of things. These types of maintenance seem to be almost inherently a case for human involvement in the process – a person observes that a machine is broken and either fixes it or replaces it.

However, the Internet of things does play a part in reducing reactive maintenance. Often, maintenance managers attempt to replace reactive maintenance with preventive or predictive maintenance. In cases where this is possible, it may also be possible to make the asset being maintained into a thing on the Internet of things.

Preventive Maintenance

The first part of this article discussed the case of a body of water – a reservoir – backed up by a dam. The problem to solve was how to detect and report when the water level was too high so that the gates would be lowered when the water threatened to spill over the dam. The solution presented was to use a sensor in the water to measure the water level and report this information to a database in the internet cloud, and then process this information to determine whether to lower the gate.

This problem turns out to be a maintenance management issue with a preventive maintenance solution. Here is one good definition of preventive maintenance:

maintenance that is proactively performed on an asset with the goal of lessening the likelihood of failure, reducing unexpected downtime, and prolonging its useful life

Applying this definition to the body of water case, the “maintenance” is lowering the gates, the “asset” is the reservoir and its gate, the “failure” is water going over the dam and damaging the countryside or, even worse, collapsing the dam. Preventing this disaster would surely “prolong its useful life.”

One key to this is knowing when to perform maintenance on an asset. In this case, there is an upper limit as to what the water level reading should be. If the water level is over this upper limit, action needs to be taken. The sensor in the reservoir can frequently check if the upper limit has been reached.

In this case, we can think of the sensor as being like a gauge. Gauges have either an upper limit, a lower limit, or both an upper and lower limit. Any gauge reading outside the bounds should trigger some preventive maintenance action or at least an inspection.

Meters, like gauges, also can trigger preventive maintenance actions. The meter reading continually increases in time with usage. A typical example of a meter is the odometer on your car, which indicates that you should change the oil every certain number of miles.

Typically, however, most people don’t use the Internet of Things to read their car odometer – they do this themselves. One place where a machine is gradually replacing human eyeballs is for reading electric meters. Instead of a human with a route reading the electric meters for each house or building, the machine periodically sends the meter reading to a database somewhere in the cloud.

A third type of Preventive Maintenance is 52-week PM scheduling. The maintenance department assembles a list of preventive maintenance activities to be performed at a certain time of year (or multiple years), often periodically. For example, the schedule could incorporate regular inspections to determine if something needs fixing. The person in charge of this scheduling then puts these preventive maintenance activities on a 365-day calendar. The calendar doesn’t have to be computer-based, but a computer-based calendar application might help schedule the periodic tasks faster.

It is hard to imagine how the Internet of things could work with 52-week PM scheduling. With 52-week PM scheduling, a human does the scheduling. There is no sensor data involved. The fact that the human scheduler might use a computer is irrelevant.

Where the Internet of Things might impact 52-week PM scheduling is in reducing the percentage of times where 52-week PM scheduling is used. When a sensor can be added to monitor the condition of the thing, it is a candidate for removal from the 52-week schedule.

Predictive Maintenance

Some maintenance management literature considers predictive maintenance to be just another category of preventive maintenance. There are some writers, however, who put predictive maintenance in its own category, and we will follow this definition. Where predictive Maintenance and preventive Maintenance are similar is that both intend to trigger maintenance before something fails. However, predictive maintenance differs from preventive maintenance in that it uses real-time data to determine when maintenance should be performed on the asset. Vibration analysis is a common example of predictive maintenance. Sensors can detect abnormal vibration, indicating that something is about to fail.

The advantage of predictive maintenance over preventive maintenance is that preventive maintenance might trigger maintenance that doesn’t yet need to be performed or miss maintenance that should have been performed. Take the example of changing the oil in a car again. For preventive maintenance, the oil should be changed every 6000 miles. Suppose the car has a bad oil leak. In this case, the car may fail before the reading of the odometer triggers an oil change.

On the other hand, suppose the car is new and without an oil leak, and the driver is a great driver. In this case, the odometer reading may trigger an oil change too soon. Resources are wasted.

With predictive maintenance, sensors can be used to continuously check the oil’s condition in the car. Maintenance is triggered only when the condition of the oil changes to an abnormal state that indicates something needs to be done.

While predictive maintenance should produce better outcomes, it has its drawbacks. The necessary equipment to perform predictive maintenance might be much more expensive than what is needed to perform preventive maintenance. It may be cost-prohibitive.

Predictive maintenance would seem to be ideally suited for the Internet of Things. The sensors that detect something is abnormal can also send that information to somewhere in the cloud.

At this point, we have discussed the Internet of Things and have demonstrated how it can be used to improve Maintenance Management. Still, IoT has not progressed quite as fast as some have predicted it would. Therefore, before looking at the future of IoT and Maintenance Management, we will first look at the roadblocks to IoT that will impact that future.

Internet of Things Roadblocks

Is the Internet of Things a success? Looking at revenue, the answer is yes. Just how big a success will depend on which numbers you are using. I found two figures for 2020:

The first estimate has global revenue for 2020 at $742 billion. What is more, this number is growing rapidly. The value in 2017 was $100 billion.

The second estimate for the same year, 2020, has global revenue at $389 billion.

That is a big difference in the 2020 estimates – from $389 billion to $742 billion! Still, even the lower number is large enough for the Internet of Things to be considered a success.

Even so, the difference in numbers between the two numbers is disconcerting. Here is a quote from the book

The Internet of Things Myth that suggests one use caution when evaluating any claim about IoT:

As a side note, in practice, “IoT” is rather meaningless – connected devices are rarely connected to the Internet. M2M [A predecessor technology to IoT – machine to machine] is slightly better – machines tend to be connected to databases (which are also machines) but a terminology that made this clearer such as “Machine-2-Database” would be more accurate. For those in the industry this does not matter overly – they are just terms used for something practitioners understand. But for the wider audience they are misleading and can result in concerns and issues.

With those caveats in mind, we’ll plunge ahead and try to answer another related question – has IoT been expanding at a rate as fast as might be expected?

While IoT has been around a lot longer than 2010, that was the first year IoT got a lot of attention from the media and general population. Various organizations started to make predictions about how many devices would be connected to the Internet in 2020. From “The Internet of Things Myth,” several of the guesses were:

Ericsson White Paper:   50 billion devices connected to the Internet
Cisco:   50 billion
Machina Research:   12 billion

The actual number of devices connected to the Internet was 11 billion in 2020. The first two estimates, the ones used by most of the press at the time, were way too high.

“The Internet of things Myth” goes on to devote most of its writing to listing reasons why IoT hasn’t taken off as fast as many have anticipated. The list below summarizes some of these roadblocks:

  1. Many modern technologies are built upon well-defined and heavily supported standards, often single standards. This is true, for example, with cell phones. Unfortunately, with IoT, there are multiple competing standards, or the existing standard is not sufficient. Also, there are many different types of things on the Internet of things. A standard that works well for one thing will not necessarily work well for another. For example, the amount of data that needs to be transmitted for one device may be multitudes different than for another device. One standard may not support both well.
  2. Supply and demand may not sync up well, or in other words, the cost of putting something on the Internet of things may be a lot higher than what people are willing to pay. Generally, the more expensive the “thing” is, the more likely someone will be willing to pay what it costs to put it on the IoT. For example, in the early days of IoT, the press gave much attention to popcorn poppers being put on IoT. It turned out this was just a novelty – the popcorn popper and what it produced wasn’t worth the cost of putting it on IoT. On the other hand, monitoring a refrigerator and its valuable contents via IoT turns out to be worth it for many people.
  3. Failure of a thing on IoT can be costly. Things on IoT tend to need more support than things that can stand alone. What happens if the company selling the “thing” goes bankrupt or decides not to support it anymore?
  4. Sometimes, for certain applications, there are better technologies to use than IoT.
  5. There are security concerns. For example, it might not be pleasant riding in a car when an IoT device in the car is hacked.

While all these roadblocks are important, I would like to discuss the last, security, in greater detail. Remember that the Internet of Things works by having devices embedded in a “thing” send data to the cloud, where the data is analyzed and processed, with perhaps an actionable message sent back to the “thing.” The basic way to hack this process is to intercept and alter or otherwise use the contents of these data streams.

This article from Dimitar Kostadinov is a nice introduction as to how this interception is done. It also discusses the extent and possible consequences of this type of traffic interception. From the article:

According to a 2020 report by a threat intelligence team called Unit 42, 98% of the 1.2 million IoT devices on corporate networks they analyzed had no capability to encrypt traffic. As a result, 57% of these IoT devices were susceptible to traffic interception and manipulation, among other things. The same report further showed that mixing IoT and IT assets on VLAN may be dangerous, as compromised employee IoT devices could spread malware onto corporate networks.

57% times 1.2 million makes for a lot of vulnerable devices. Why are such a small percentage of devices protected? Protecting these devices from hacking can be both difficult and expensive.

The consequences of these vulnerabilities can be deadly. We have already talked about how a human can be a “thing” on the Internet of Things. The article The 5 Worst Examples of IoT Hacking and Vulnerabilities in Recorded History includes two frightening stories about health devices. In one case, a hacker could deplete the battery of a cardiac device. In another, a hacker could produce misleading readings from a baby heart monitor.

The same article cites the case of a vehicle that could be hacked. Jeep had to recall 1.4 million vehicles because researchers proved hackers could kill the engine or disable the breaks.

Smartphones are not immune to hacking either. From Forbes, discussing a 2021 incident:

A vulnerability in a chip manufactured by $60 billion market cap Taiwanese tech giant MediaTek left a third of all of the world’s smartphones and Internet of things devices open to remote snooping of phone calls and spying via the device microphone, researchers have claimed.

Fortunately, these vulnerabilities were addressed quickly by October of 2021. Otherwise, that would have been a lot of phone owners who might have been snooped upon.

These examples of security breaches paint a picture of a dangerous technology. IoT can be dangerous, but the same is true of other new technologies in the past. I remember not too many years ago that my personal computer would get a virus about every six months even though I had virus protection from a well-regarded company. That, knock on wood, just hasn’t happened in the last few years. Virus protection seems to be getting better. The same will hold true for security on the Internet of Things.

Why should this be? Economics again is part of the answer. While providing better security for things on the Internet of Things can be expensive, not providing security can be even more expensive. Suppose Jeep had not recalled those 1.4 million vehicles and fixed the security problem. The cost of lawsuits and lost sales would have overwhelmed the cost of fixing the problem.

There are other parts to the answer. We humans keep learning things, and once that knowledge becomes common knowledge, it becomes free. What one of us learns about improving security, the rest of us are free to apply to other situations.

Another part of the answer is that the sheer size of the IoT industry almost guarantees that security will be improved. The larger the industry, the more people will be involved in providing security. You and I could get a certificate in IoT security if we wanted to.

Another reason for optimism about security is that IoT will evolve in ways that make things inherently safer and more secure. Further, if IoT is not safe enough for certain applications, there are similar but competing technologies that might work better. Take the case of self-driving cars, where IoT may or may not be needed. One of the chapters in the book “The Internet of Things Myth” is devoted to “Autonomous driving” – where the car drives itself without human intervention. The second paragraph in this chapter hints that IoT connectivity for cars may not be needed for many cases:

For some time, autonomous cars such as those from Google and Tesla have been driving themselves around California without much connectivity. But now we are told that ubiquitous connectivity delivering ultra-high data rate and millisecond latency linked to a centralized control system is critical to the future of the autonomous car. What has changed to require this, or is this not actually a requirement at all, rather than the IoT industry desperately seeking a use case?

The authors’ answer is that sometimes IoT connectivity is needed or is at least useful – software updates and maps need to be downloaded to an autonomous vehicle. This is a somewhat intermittent activity – the downloads might be needed once a day or less often. As we saw in the case of the Jeep recall, fixing a security breach here might be expensive but doable.

On the other hand, having IoT completely control every movement of the car would require vast amounts of data updated continuously. Hacking that data might have more serious consequences. To take one example, suppose you wanted your self-driving car to avoid a collision if the car ahead of it slowed down suddenly. The IoT solution would involve continuously monitoring the speed of both cars and the other conditions of the road. There is another solution not involving the IoT that might be easier, cheaper, and more secure to implement. This is called vehicle-to-vehicle (V2V) communications. The vehicle in front signals to the car behind it that it is braking. The car behind uses this information to know that it should slow down as well. None of this involves sending data to the IoT “cloud,” making the process simpler and the car less hackable.

To implement vehicle-to-vehicle communication probably means that more computer brainpower is needed at the local level – at the level of the car – to displace what is lost from the IoT cloud. “IoT vs Edge Computing: What’s the difference?” goes into a general discussion of how one technology might allow something like this to work. To briefly summarize, for the Internet of Things, the IoT devices – sensors + their accouterments – are efficient single-purpose devices that can’t do much data processing on their own. Instead, the data that they accumulate must be sent to the cloud to be processed, with the results sent back from the cloud and used to perform some action.

The alternative presented by this article is called “Edge Computing. “The Edge devices have a lot more computing power than the IoT devices, are located near the “thing” that they are monitoring, and don’t need to send their data to the cloud for processing. Instead, they can do the processing themselves and then act on the results. So while the Edge devices can still be hacked, it would be less easy than with IoT devices.

“Edge Computing” isn’t something that will happen far in the future. It is newer than IoT, but it is becoming more viable because the computers available for “Edge Computing” are becoming more powerful. Of course, security isn’t the only reason Edge computing might be preferred to IoT – Edge computing can be faster, more powerful, and reliable – but it is a major consideration.

We should note that when it comes to safety, IoT can be used for good as well as evil. For example, a hacker can compromise the safety of a dam by sending harmful messages to the dam, but IoT, when used properly, can improve the security of the dam, as was shown earlier in this paper. In our example, we used the example of an IoT message sent to the dam to lower the gates to prevent a breach of the dam. Besides breaches, dams can also have structural flaws and dangerous obstructions, and IoT can be used to detect these flaws and obstructions.

So, there are reasons to be optimistic that security problems for the Internet of Things can be resolved with the security of things actually improved. Even so, it never pays to be too blasé about IoT security. When the question “can nuclear weapons be on the internet of things” is asked, this article answers “yes:”

Many Americans have become aware of the so-called “Internet of things”—the vast but largely unnoticed number of smart appliances and objects that connect to the Internet. A few years ago, the National Security Agency discovered suspicious electronic emissions coming from a sensitive facility—a significant breach of security. After a long investigation, the culprit turned out to be a soda machine, communicating to its vendor over the Internet that it needed to be restocked. What happens when the Internet of things includes nuclear weapons? An odd question, perhaps, but one that the Air Force Scientific Advisory Board is asking. New nuclear weapons systems “will be much more like all systems today, network connected,” the head of the board told reporters. “They’ll be cyber-enabled”—suggesting that new systems will be as much a part of the Internet of things as thermostats and refrigerators.

That is a little scary.

Back to the question, is the Internet of Things a success? I may have made it seem that the authors of the book “The Internet of Things Myth” were pessimistic about the IoT industry. On the contrary, the authors are bullish. One of their main points was that the optimistic predictions about IoT may still come true, but the date may be delayed – instead of 50 billion connected devices by the year 2020, it may not happen until 2030. In the long run, does that really matter?

Another of their points is that governments and corporations need to learn from the past. To hasten the progress of IoT, a lot more work and cooperation needs to happen to establish workable and universal standards. IoT developers need to think carefully about what should and should not be on the Internet of Things. The authors have a lot more advice for expediting the future of IoT that I will not rehash here, but I will leave this topic with their final advice, “above all, learn from previous mistakes.”

The Future of the Internet of Things

In the introduction, I promised not to engage in science fiction-type examples. Yet, even when we talk about the Internet of Things in the most mundane and conservative ways, the effect on our lives will likely be astonishing.

  1. As an industry, the revenue growth for IoT will be enormous. To illustrate what I mean, compare IoT to another large industry – the smartphone industry. Statista estimates global smartphone sales in 2020 to be 409.1 billion dollars. This is the same website that provided us with an estimate of global IoT revenues in 2020 of $389 billion, not that much smaller than that for smartphones.

    But smartphones are an industry nearing maturity. Statista, in the same article, estimates that in 2019 40% of the world had access to smartphones and that that percentage had not been going up for several years. On the other hand, IoT is nowhere near maturity, with some kinks remaining to be worked out before it can really boom. Once it reaches this point, it has much more room to grow than the smartphone industry.

  2. IoT will affect many industries (This isn’t much of a prediction because it is already true. Let’s just say it will become “truer” as time goes on). One book I mentioned previously, Internet of Things for Architects: Architecting IoT solutions by implementing sensors, communication infrastructure, edge computing, analytics, and security, lists a number of industries affected by IoT plus their use cases:
  3. Industrial and Manufacturing – impacts preventive maintenance, throughput, energy savings, safety.
    • Consumer – from coffee pots to thermostats to Alexa assistants to Fitbits, etcetera.
    • Retail, financial, and marketing – targeted advertising, asset tracking
    • Healthcare – in-home patient and elderly care, predictive and preventive healthcare
    • Transportation and Logistics – asset tracking, preventive maintenance of vehicle fleet
    • Agricultural – irrigation, livestock monitoring
    • Energy – Monitoring oil rigs, solar panels, nuclear facilities, electricity usage

      I’ve left out several industries and many of the use cases listed in this book. Again, as mentioned, this book is a good place to start your IoT education.
  4. The uptake of IoT by industry and use cases will vary greatly. Often, we see a newer technology replacing an older one – cars replacing horses and buggies or cell phones replacing landlines – but it doesn’t have to be this way. For example, there may be a place for smart doorbells, but unless the government mandates them, we will still see homes with standalone doorbells, door knockers, and even doors that you just have to pound on to get an answer. It is easier to predict that a connected embeddable medical device, if it provides much greater value than the existing technology, displaces that existing technology.
  5. Beginning with the Luddites, many people have worried that machines will replace human workers. So far, that hasn’t happened, and a lot of economists predict it won’t. Still, IoT will change a lot of the patterns of employment. Suppose, for example, that governments mandate that electric meters are read via IoT rather than by human beings. If those mandates are seriously implemented, that eliminates the need for an entire group of workers.
  6. There will be a lot of contentious debates about certain uses of IoT, especially when the government is involved. The key argument will be how much control a government should have over your life. An example is the use of a “kill switch” to allow law enforcement to turn your car off.
  7. Maintenance Management will be an area of work heavily impacted by the Internet of Things. Systems will rely less on workers to detect potential failures and more on devices (sensors) that can monitor equipment continuously and send data back to the Internet cloud. That alone will have a profound impact on Maintenance Management. The next step of IoT will be to get to the point where it is commonplace for devices on the IoT to act independently of human beings when determining what and how something should be fixed.

None of these predictions is new or even startling, given what is already happening. However, there are changes ahead that are not quite so easy to predict. What is amazing to think about is that in the next couple of decades, the Internet of Things may have as large or even larger an impact as what smartphones or the Internet itself have had in the last two decades. The science fiction author Arthur C. Clark has a famous quote “Any sufficiently advanced technology is indistinguishable from magic.” Will the Internet of Things be the first technology to reach that point? A lot of us may live long enough to find out.

Related Articles



  • Aaron Tobiason

    Aaron Tobiason is a retired computer programmer old enough to have programmed dBase II on CP/M machines. He started off doing contract programming on a multitude of subjects, but for the last twenty-five years has been programming Computer Maintenance Management Systems.

Subscribe to Our Newsletter

Share This Article


Recent Posts