Elevator predictive maintenance: from words to reality

How predictive maintenance can solve the global elevator problem, if executed with the right product vision.

Our previous article explored how the elevator maintenance business model is in bad shape:

• It revolves around an outdated means vs. ends model in which the customers purchase maintenance visits and breakdown response.
• This model encourages an inefficient productivity hunt by service providers, driving up breakdown rates and reducing the quality of service.
• These productivity gains then entail structural churn of the portfolio units and a decrease in maintenance prices — we call this process the commoditization of the market.
• This commoditization is barely counter-balanced by the new elevators installed and converted to maintenance services, as too many old lifts churn to local players.
• The COVID-19 crisis is further endangering the model, as new equipment installations have come to a temporary halt and regulation softening might accelerate.

We looked specifically at the situation of the leading four OEMs and concluded that

• while they excel at product innovation for new elevators, this necessary defensive move is not enough to reverse the trend, and
• their predictive maintenance efforts seem to be stuck at the marketing stage. This is due to their manufacturing culture, with innovation rooted in their factories. They lack a retrofit innovation culture and a software culture.

In this article, we look at the change that is needed to derail and even reverse the downturn and why this relies on high technology. We offer advice on how to implement such technology, with the example of our own company, and conclude by forecasting three possible scenarios for the industry’s future.

What is quality?

In the elevator service industry, quality is the sum of

1. technical quality: performance of the service, e.g., the breakdown rate; and
2. perceived quality: customer engagement in a satisfactory relationship with the service provider, e.g., communication with customers during a breakdown.

The first is necessary, but not sufficient; the second on its own is useless. The two should be built on top of each other; the best communication will not help if the elevators are always down, and perfect maintenance is worthless if customers aren’t aware of it.

Technical quality

Five indicators assess the technical performance of elevator maintenance:

1. Breakdowns: their frequency, usually a total per year.
2. Uptime rate: the total time when the machine was working, as opposed to not working, in %.
3. Callbacks: the customers’ complaints, which can relate to a breakdown or any other visible problem, such as a broken light or a broken button.
4. Asset preservation: whether the elevator is well maintained and its organs frequently repaired to guarantee the longest possible lifetime vs. a costly overhaul.
5. Compliance: whether the elevator is complying with all local regulations.

Most elevator companies have been focused only on the third factor — the callbacks — as it’s the most visible.

Market players rarely log breakdowns vs. callbacks and are therefore not able to calculate their exact uptime rate.

Asset preservation is considered the other way: how can we sell the earliest possible overhaul? Compliance, hopefully, is not overlooked.

These indicators, which would seem useless in a means vs. ends sales model, will tell you whether you are on track to reverse the trend.

Perceived quality

What customers can see and feel is perceived quality:

1. Ride quality: how comfortable is the elevator ride, from waiting time to lighting and noise?
2. Information: how easily accessible, clearly understandable, and transparent is the information provided?
3. Brand: how does the brand experience generate trust toward the service provider?

The market standards don’t rank higher when looking at perceived quality. In most buildings, tenants can wait months or years before a faulty floor button gets changed.

Information is technical, opaque, and poorly accessible — even professional property managers systematically complain about having a hard time getting precise KPI reporting or a digital dashboard.

At uptime, when we asked our customers for the precise breakdowns’ performance they had had with their previous service providers, none could actually answer.

Big OEMs have focused their efforts on developing their manufacturer brand, not their service brand. Maintenance is marketed as a human-first, proximity service, but what makes people churn? Breakdowns and lack of information.

To reverse the trend, perceived quality is not a nice-to-have. What’s the point of solving technical problems if your customer doesn’t know about or understand it? Customer expectations have risen in the past decades; the job not only needs to be done, it also needs to be marketed instantaneously.

Doesn’t quality decrease margins?

In the historical model, delivering higher quality is directly associated with increased costs: more technician time, more support teams, and more marketing costs. Unless a contract price surge compensates for it, margins go down.

For productivity-driven management, the choice is quick and easy: increased quality is guaranteed to increase costs but is uncertain to raise prices. So, they just don’t touch it.

Well, that is the old industrial universe, but now, software is eating the world.

High technology can reignite portfolio growth

Software innovation in the elevator maintenance industry can

• help provide around-the-clock, precise, and transparent data for customers, increasing perceived quality and the brand, and
• help reshape the maintenance delivery model and thus reduce breakdowns and other failures, increasing technical quality and decreasing costs.

Tangible changes in both real and perceived quality reduce churn and allow for better differentiation and thus higher prices. Eventually, high technology enables companies to renew their organic portfolio growth while reducing costs.

We will not deep-dive into the construction of perceived quality based on modern software and B2C branded experiences, as there are plenty of startup benchmarks around the world. Let’s instead focus on the tech that reshapes the delivery model: predictive maintenance.

In the context of elevators, predictive maintenance combines precise, real-time sensor data with field information to determine the best time to perform specific maintenance operations, thus improving the quality while controlling the costs.

Predictive maintenance starts with data

Reshaping the delivery model and avoiding breakdowns is possible only with the right dataset. Predictive elevator maintenance cannot happen without two things:

1. Field data. What did the technician do? Did he tighten a screw? Did he change a spare-part? Did he observe irregular wear of a component?

2. Elevator data. How much did the elevator travel during the day? How many times did its doors open? Is the door-opening cycle stable? Do the electronics warn of specific statuses or failure?

It is impossible to predict whether a component will need to be adjusted, repaired, or changed without knowing when it was last fixed and how many times it has functioned since.

Could you picture car maintenance without mileage counters and knowledge of past interventions? Well, that is precisely how elevator maintenance operates today.

Half the necessary data doesn’t even exist

Maintenance service — focused on compliance with regulations — has not helped to collect any data on the installed base:

• The how-to for solving failures, adjusting various components together, and avoiding breakdowns is in the heads of the service technicians. They only share this knowledge-base orally.
• There is no tracking of the actions these technicians perform on-site. Until recently, most maintenance companies still lived by paper-and-pen reporting. Digital reporting today is not normalized: a text field saying “I fixed the failure” does not create usable data.
• There is not even any tracking of the components: every elevator is an assembly of multiple manufacturers, and most of its elements change over the years. Nobody keeps track.

Normalizing — with the right software applications — every aspect of the installed base and every action taken by a technician on-site is the first requisite of predictive maintenance.

Elevator data exists but needs processing

Lifts already have dozens of sensors: exact position sensors in the shaft, door sensors, door lock sensors, temperature sensors on the engines, and so on. Lifts also have a number-of-trips counter, but it is not readable without a specific tool and therefore never used.

Technicians never replace a component in advance after taking into account its mileage — they don’t know the mileage: they wait until it breaks down.

The most obvious way of building IoT for elevators is thus not by adding sensors, but by obtaining and processing the existing data.

Elevator IoT: technical choices are vital for quick ROI

Data is in the controller

The main electronic board — or controller — of an elevator gathers all this pre-existing data. With the right tools, it is possible to access the logs of the machine and therefore measure

• the traffic, such as the number of trips, door openings, and floor usage;
• the functioning of the components, the engine starts and stops, and the cabin door cycles; and
• the sensor data and the statuses computed by the controller board.

We thus believe that the shortest way to build tons of usable data is by connecting digitally to controllers of any brand. This is precisely the hardware that uptime has developed.

uptime’s universal controller-IoT card

Vibration analysis is a long shot

Some predictive maintenance initiatives in the industry have focused on installing additional hardware — specifically, accelerometers — to capture in-shaft vibrations. In the past, vibration analysis has had some successful use-cases, such as

• in the elevator industry, to benchmark the ride quality of a newly installed elevator, and
• in the manufacturing sector, to help maintain rotating engines.

However, both these use-cases operate in predefined and controlled environments, which is not the case with the elevator installed base. The park is aging and subject to continued external actions from users and technicians.

As a counter-example, the car industry could have a bright future with vibration-based predictive maintenance. It has a long-standing process for noise and vibration data — NVH; every part of a car has a manufacturer’s NVH signature when leaving the factory, and a car model’s assembly remains the same over the years. None of this applies to elevators.

Vibration analysis can be quick and cheap to develop and deploy but extremely slow and expensive for providing tangible insights compared to controller data.

Don’t OEMs already have controller data?

An OEM assembles mechanical parts from various manufacturers and conceives the electronics: by definition, it has access to its own controllers’ data. Neil Green, Otis’ Chief Digital Officer, told InformationWeek that Otis has collected daily data from over 300,000 of its elevators since the 1980s via their Remote Elevator Monitoring (REM) tool. In other words, Otis invented the IoT for lifts.

So, what went wrong?

1. Data structure. Receiving data once a day about an elevator’s current status is very different from accumulating, in real-time, all the events from a controller to feed into an AI model.
2. Data usage. IoT data, if not associated with relevant technician data and not targeted at reshaping maintenance methods, is useless. Data alone will not produce any ROI.
3. Universality. OEMs have access to their controllers, but they maintain dozens of makes. This might be one of the reasons Otis says it installed only 300k REMs out of a 2M unit portfolio. It is hard to change field methods when those changes can’t apply to all serviced elevators.

IoT elevator data alone is useless; a predictive AI with access to it won’t generate any tangible impact. The whole service delivery needs to be data-driven. It is like installing a 2020 Formula 1 computer into a 1990s mainstream car with no electronics; it’s nonsense.

Data-driven maintenance is Phase 0

If service is to change from being technician-oriented to being technology-enabled, it is necessary to systematize field operations before any predictive maintenance can happen.

Know your portfolio

To perform first-level analytics and feed a predictive algorithm properly, knowing the installed material is essential. A component’s cycles of wear are different by make, model, and environment, and referencing and updating the main organs — at least the controller, the door operator, and the engine — of the installed base is the first task. It sounds obvious, but today, most companies do not have proper data on their portfolio.

Know your failures and field actions

Second, you need to normalize all failures. Failures cannot be just callbacks in a call-center database. Their precise qualification will feed the algorithms:

• In what exact state was the elevator when the technician arrived?
• On which components did he act and how?
• What was the status when the technician left?

Third, actions performed by technicians outside of failure situations have to be recorded precisely:

• Exactly what procedures — visual check, operation test, repair, lubrication, adjustment, replacement, wear-level check…
• Exactly which components.

Out of all the data necessary for predictive maintenance, our experience shows that field data can be the hardest to generate. Choosing the right level of abstraction against the thousands of different components and environments, creating the right software user experience, and efficiently changing the technician work culture are the main challenges.

Measure

This initial dataset can already provide some analytics and enhance field performance; now you can investigate useful metrics, such as

• % callbacks that are not breakdowns,
• uptime rate,
• breakdown rate by material or component type,
• breakdown rate after a specific field action,
• first-time fix rate on a failure type, and
• success rate of various failure responses.

And you can start to correlate field actions and failure patterns.

Notice that this hasn’t yet required any IoT — just the right software with the right user experience for the technicians. If we add elevator data, we’ll be able to tailor operations to each lift: we enter predictive, condition-based maintenance.

Now we can plug in the Formula 1 computer. But wait — what for?

The right product vision wins

Product is focus

The great thing about modern technology is that it can do almost anything. The downside is that nearly every one of those possibilities is entirely worthless, and large companies that do not have a software innovation culture struggle to grasp this.

Successful technology startups do not win by the depth of their resources, but by the precision of their focus.

Technology-driven quality kills costs all day long

The end-game is simple: organic portfolio growth, both in volume of serviced elevators and in profit generated per contract.

However, the critical path is often blurred by productivity thinking: shouldn’t one focus on predictive maintenance’s impact on costs first, as they are more visible than the indirect outcomes of a higher quality of service?

Focus on quality first, and costs will reduce themselves:

• If your first-time fix rate increases and your breakdown rate decreases, this means fewer field hours for tackling callbacks.
• If your maintenance operations are data-driven and standardized, it becomes easy to lower the skillset barrier for value-added interventions.
• If your perceived quality is high and the information provided on the service is abundant, it becomes easier to sell necessary additional repairs.
• Once the right quality is reached, it becomes easy to raise the average contract price and, therefore, the margins.

Predictive maintenance is not an add-on. It is a reconstruction.

Elevator OEMs today have a simplistic vision of predictive maintenance in their industry: plug in an IoT device, gather data, anticipate breakdowns before they happen, and send a technician. This cycle is seen only as an addition to what the technician already performs:

• Mostly useless compliance visits. We call them “ghost visits.”
• Breakdown response: receive a callback, get to the building, reboot the elevator, and leave.

Not only is this add-on vision hard to achieve, as pure IoT data is rarely sufficient to anticipate breakdowns, but it entirely misses the point. It generates additional unplanned field hours (the most expensive ones) instead of reducing them.

The right product vision is a reconstruction of the service delivery from scratch:

• Rebuild the technician user experience to gather normalized field data and to standardize field actions.
• Within local regulation boundaries, avoid unnecessary field hours with a differentiated (high-value or compliance-only) visit model.
• Focus predictive maintenance efforts on early breakdown avoidance by applying the right preventive actions: transform expensive curative field hours into cheaper, predictable planned hours.

The two phases of predictive elevator maintenance

The warm-up: enhanced breakdown response

Once you have gone through Phase 0 — gathering field data — and plugged in your IoT device, you can start deploying smart failure response. We have constructed Phase 1 around three core features:

1. Breakdown detection: anticipate customers’ callbacks by detecting whether the elevator is up or down in real-time.
2. Callback qualification: is the callback a real breakdown or a less urgent matter?
3. Technician guidance: depending on the elevator status, the visual checks by the technician, and the error codes provided by the controller, what are the right tasks to solve the root cause of the failure?

Building this first set of features will already yield significant impact:

• Detection builds perceived quality: customers crave instant notifications of failures instead of having to call an emergency number.
• Callback qualification saves field costs: the ability to distinguish real urgencies from problems that can be solved over a more extended timeframe allows transformation of unplanned hours into cheaper, planned ones.
• Guidance saves field costs and increases technical quality: precise elevator data, in addition to technician contextual help, raises the first-time fix rate, reduces the field hours needed, and increases satisfaction.

The real stuff: breakdown reduction

For Phase 2, we’ll describe how, at uptime, we are quickly building breakdown-reduction features through predictive maintenance. The process described below is part of our filed patents on the topic and can be summarized in three parts:

1. Identify key preventive tasks.
2. Compute their correct realization timing.
3. Deploy them through a dynamic visit model.

Identify key preventive tasks

To define the correct tasks that avoid breakdowns months in advance, the first step is an in-depth analysis of the failures:

• Failure profile by component and frequency.
• Understanding of the root causes of these failures and their associated curative actions.
• Determining the correct preventive measures to entirely avoid the root cause.

This creates a dynamic list of preventive tasks per component associated with a failure profile, which can already be achieved with data-driven operations and technical expertise.

Identify when to perform these high-value actions

Now we need the Formula 1 computer: the IoT data and the AI to generate timely recommendations. The output we need is the correct timing for preventive tasks:

• Early enough to avoid a breakdown.
• Late enough to minimize field interventions and costs.

The data we combine to generate the most tailored timing includes

• field checks: initial wear check, installation date, and regular wear checks;
• IoT data: traffic statistics and component usage; and
• context: breakdown patterns of the specific component and the specific elevator.

Implement a dynamic visit model

The last step looks like the easiest: push these tasks to the high-value visit checklist of our technicians. Each maintenance visit is different to the previous one and tailored to each elevator — in other words, condition-based maintenance.

However, to achieve successful results, user-friendly guidance on how to act is required. Indeed, if the implementation is not standardized, the performance of the predictive maintenance campaigns will vary greatly depending on the technician profiles.

Cultural change is needed to both empower the technician and, at the same time, reduce the skillset barrier for value-added tasks.

In this new model, technicians are not alone in their cars anymore, running from visit to visit (or, in a high breakdown–rate context, from reboot to reboot), with their competences largely under-utilized. They participate in a broad program of quality enhancement that requires from them discipline and precise reporting (data-driven operations) and strong involvement in reducing the root causes of breakdowns (dynamic visits).

Incumbents cannot solve predictive maintenance alone

Historical actors must rely on technology partners

Elevator OEMs or SMEs have a classic engineering culture, not a software innovation culture. To reconstruct the maintenance model, one needs

• disruptive integrated technology (hardware, software, and AI),
• retrofitting and universal innovation, and
• widespread cultural changes in the maintenance organization.

The first two requisites can only happen with partners that have the right disruptive, focused, and fast innovation culture — mostly from the startup world. As described by Christensen in 1997 in The Innovator’s Dilemma, almost no global company can disrupt itself. The recent history of the elevator industry, with initiatives that are add-ons to current processes, focus on the OEM’s make only, or do not yield significant business outcomes, confirms it.

Technology partners need a laboratory portfolio

Why do innovation programs by the OEMs — in partnership with tech companies like IBM or Microsoft — not seem to disrupt the industry? Why have independent material suppliers from the elevator industry not yet commercialized results-oriented IoT?

As demonstrated, cold IoT data doesn’t solve problems. Field data from technicians is needed, together with a reconstructed service delivery with dynamic visits. It is impossible to create a compelling predictive maintenance product for elevators without operating a maintenance portfolio that serves as an agile test environment.

We understood this necessity early-on at uptime, and today we might be the only elevator tech startup with a test portfolio.

Digitalize or commoditize. What’s your call?

How will the market evolve? We believe there are three possible paths.

Scenario 1: full commoditization

In this scenario, the OEMs won’t be able to reverse the trend; their service market share will continue to slide over to local independent service providers. The market will become fragmented, and its total value will stabilize or even decrease.

A standalone player or two will emerge above the others, thanks to technology. The OEMs will focus on building qualitative equipment and increasing their manufacturing margins, but they will lose greatly in market capitalization.

Scenario 2: OEMs become tech-enabled and service-driven

One or multiple OEMs will achieve a turnaround with the appropriate partners and rebuild their service operations around three changes:

• Retrofit-oriented, universal innovation.
• Real predictive maintenance in reconstructed service operations.
• Proper technical quality and perceived quality.

Those OEMs will be renewed with organic portfolio growth and take significant market share from the others and from independent providers. There will be fewer actors, and the market’s value will expand.

The winning OEMs — now technology-enabled rather than technician-driven — will be seen as technology companies relying on a valuable base of long-term contracts, and their market capitalizations will rocket.

Scenario 3: independents dominate with the right toolset

SMEs will onboard new technology from the right players, such as uptime, and turbo-charge their growth. They will digitalize faster than the OEMs and focus efficiently on quality of service based on predictive maintenance.

The transfer of contracts from the OEMs will accelerate, and new service actors will be frequently created in all geographies, building on top of technology providers. There will also be some regional build-ups helped by private equity.

As in the first scenario, the OEMs will focus on product development, increase new equipment margins, and progressively leave the service business. Overall, there will be more players, and the market’s value will stabilize or even grow.

Uptime

At uptime, over the last 3.5 years and with €15M of funding, we are solving predictive maintenance for elevators. We have built the technological stack from the IoT to operational apps and customer dashboards. We are rapidly developing the AI and generating solid results in our organically-grown maintenance portfolio.