Algorithm-based data

analytics for lifts

How data analytics elevate your

lift business to the next level of
effective efficiency

White paper

Abstract
Operators generally lack the necessary decision-making skillset to determine optimum maintenance activity. However,
lift availability restrictions and safety issues can be prevented if clear operating parameters are monitored, analysed and
objectively interpreted. Alongside this, OEMs are investing in digital solutions to reduce physical maintenance visits, whilst
streamlining operational processes and workflow. The TÜV SÜD Lift Manager solution offers lift operators, OEMs and
service providers a unique solution to this problem.

TÜV SÜD
Contents
Pains and gains of the lift business and the benefits of digital transformation 3

The Lift Manager – the universal digital solution for operators, OEMs and service providers 5

The Lift Manager solution at a glance 10

About TÜV SÜD 10

About the TÜV SÜD expert

Dr Rolf M. Zöllner
Head of Business Development for Lifts, Cranes and Conveyors
As TÜV SÜD’s Business Line Manager in Smart Lifts Division for the international business
and Head of Business Development in Lifts, Cranes and Conveyors Business Unit in
Germany, Rolf drives digital transformation of technologies, services and contents across
TÜV SÜD businesses. Rolf is a trained mechanical engineer. He also has deep experience in
functional safety, testing, inspection and certification services accumulated through a broad
spectrum of industrial applications in his 14-year career at TÜV SÜD.

Ahamed Moosa
Senior Data Analyst, Digital Service, Singapore
As a data analytics professional with a master’s degree in Enterprise Business Analytics,
Ahamed has 9 years of experience across various industries with the focus on forecasting,
rule engines, machine learning, optimisation and business intelligence. He started his
TÜV SÜD career as a Senior Data Analyst in the Smart Lift Division. He is responsible for
the development and implementation of preventive maintenance and anomaly detection
for elevators.

2 Algorithm-based data analytics for lifts | TÜV SÜD

Pains and gains of the lift business and the benefits
of digital transformation
Lack of transparency and growing
technological complexity are
increasingly challenging lift operators,
original equipment manufacturers
(OEMs) and service providers. Why is
that? Relying on conventional failure
diagnosis and reactive maintenance
regimes to maintain safety and
availability is becoming more difficult
with every innovation cycle.

Starting with the operator

At the moment, in the event of failure operators lack sound and objective more high-rises are being constructed
the lift or its control system does not decision-making skills required to to address urbanisation needs. While
automatically inform the operator. determine the case-related optimum for smaller buildings the lift is a
Instead, it is common for informal maintenance activity, to avoid lift and question of comfort, in taller buildings
information flows to be relied upon, component failures and downtime. it is an essential operational element.
for instance via a message to the
facility manager from tenants or Of course, safety is another essential To conclude the operator’s view - it is
building users. Once the information concern for the operator. Issues obvious that availability restrictions,
is finally received and maintenance can even arise when sufficiently as well as safety issues through
activities are proposed to the building trained and experienced technical unexpected lift incidents and quality
owner, they are unable to validate professionals are under pressure to do issues in maintenance activities,
these activities in terms of necessity, their job within tight time schedules. can be prevented if clear operating
scope, cost appropriateness, Is that the right way to maintain lifts parameters are monitored, analysed
technical effectiveness and economic properly? Certainly not, especially and objectively interpreted.
efficiency. Generally speaking, as buildings are becoming taller and

What is the situation on the OEM and service provider side?

The key challenges for OEMs contracts guarantee high availability in digital solutions and seek to recover
and service providers are high of lift operation and they must fulfil these costs by reducing physical
maintenance service costs and a these contractual obligations while maintenance visits and streamlining
lack of trained and experienced simultaneously realising economic operational processes and workflow.
personnel. Moreover, OEMs’ service benefits. To achieve this, OEMs invest

TÜV SÜD | Algorithm-based data analytics for lifts 3

Operator and OEM – complementary parties but common pains
What are the lessons to be learned and downtimes. ■
Lack of skilled and experienced lift
from that? Overall, operators, OEMs ■
Missing systematic and real-time maintenance professionals.
and service providers face the information about a lift’s operational ■
High costs of physical maintenance
following common challenges: status and condition. and service in the face of growing
■
Predominating reactive technological complexity and
■
Lack of transparency of maintenance regimes or expensive narrowing time frames.
maintenance quality. preventive maintenance, instead ■
Lack of objectiveness, reliability
■
Lack of objective and reliable of proactive and need-oriented and validity in the interpretation
decision-making, as well as valid actions and planning. of relevant key performance
indicators for operational planning. ■
Heavy reliance on maintenance indicators.
■
Unpredictable occurrence, activities with random sampling
frequency and duration of incidents and visual inspection.

TÜV SÜD expertise turns pains into gains

How can we turn these pains into approaches, such as probabilistic trends can be observed for offering an
gains? This question can be resolved networks, algorithms and Artificial integrated solution:
through the systematic combination of Intelligence (AI).
the following disciplines: 3. Lift domain know-how. Expertise 1. On the one hand, solution
in the normative requirements, providers can invest in upskilling
1. The Internet of Things (IoT). technology, technical design staff and broadening their
Expertise in connecting physical and functions, as well as the competencies, which is costly and
objects through sensors, collecting operational processes. time-consuming.
data on their status, condition and 2. On the other hand, solution
utilisation, processing data and However, there is hardly any company providers can enter into strategic
presenting them in ready-to-use that can cover all of these disciplines partnerships that systematically
graphical form. consistently at a high level. The package together the individual
2. Data analytics & data science. core competencies of companies core competencies. This is exactly
Expertise in generating knowledge that offer solutions to operators and the approach TÜV SÜD took
and information from collected OEMs usually address only one of the to make their solution, the Lift
data using scientific methods and disciplines outlined above. Thus, two Manager, a reality.

4 Algorithm-based data analytics for lifts | TÜV SÜD

TÜV SÜD Lift Manager – the universal digital
solution for operators, OEMs and service providers
The Lift Manager is the result of defects etc. entire lift portfolio into a single IoT
leveraging the synergy systematically 3. Predictive analytics and platform solution. Thanks to the non-
in three areas - IoT, data analytics maintenance, e.g. exceeding invasive retrofits of a sensor node on
and data science, and TÜV SÜD’s door vibrations before a door the top of the lift car, the solution can
comprehensive lift domain expertise. roller breaks. be applied to both new and existing
lifts, regardless of their brand and
Our solution provides three crucial The realisation of these capabilities model. In addition, the Lift Manager’s
capabilities that lift operators and through the Lift Manager offers edge computing capabilities and its
OEMs are looking for: lift operators, OEMs and service configurable standard API enable
providers a unique solution that is easy integration with other platform
1. Remote operation monitoring, e.g. manufacturer-independent. This solutions, to give the client a high level
trips, door movements, levelling etc. approach offers all stakeholders a of flexibility.
2. Anomaly detection and alerts, e.g. strong advantage, both technically
emergency stops, wear and tear, and analytically, when integrating an

The brain of the Lift Manager

The following sections illustrate The analytic engine of the Lift includes a “dashboard” from which
the extraordinary approach and Manager is a highly modular and the analytic engine’s results can be
features that TÜV SÜD applied when configurable software system for integrated with and displayed on any
developing the Lift Manager’s “brain” delivering condition monitoring and existing systems.
– the analytic engine, its capabilities predictive maintenance results for lifts
and results. into IoT platforms. The standard setup

TÜV SÜD | Algorithm-based data analytics for lifts 5

Take advantage of new digital technologies
Today, lift maintenance is mainly Cost savings:
physical. The quality and efficiency of ■
Minimising wear and tear of
the maintenance regime is essentially The Lift Manager is lifts and components with real-
determined by the skills and the designed to resolve time monitoring and predictive
experience of the lift technicians these limiting factors maintenance system.
and the time spent on diagnosis. ■
Using algorithms to reduce life
The Lift Manager is designed to
through continuous cycle costs by ensuring high-level
resolve these limiting factors through monitoring of the lift maintenance to avoid premature
continuous monitoring of the lift operation and transform refurbishment or unnecessary
operation and transform such data replacement works.
to actionable insights. Consequently,
such data to actionable
all lift stakeholders can realise the insights. Process optimisation:
benefits and opportunities of higher ■
Optimising of maintenance
efficiency and productivity, while planning, scheduling and regimes.
increasing the safety and availability Predictive Maintenance ■
Supporting the integration of
of the entire lift portfolio. The use of “Lift Manager detects and evaluates smart technologies for smooth
customised sensors, data-crunching an abnormal behaviour or change and reliable lift management and
edge computing, highly available before an anomaly detection threshold operations.
cloud services and sophisticated is finally reached.”
AI-backend data analytics are Connection of lifts under an
fundamental differentiators of the Lift It is the application of data independent interface:
Manager. Operators, OEMs and lift interpretation to detect and predict ■
The Lift Manager is a drop-in
service providers are looking for: degradation processes before a solution that can be retrofitted on
failure or defect occurs. It is the most any lift brand or model.
Operation Monitoring complex and advanced application ■
Single point of access, as all assets
“Lift Manager’s operation monitoring of the Lift Manager and offers the are connected to the IoT platform
collects all necessary lift and service greatest benefit for the end customer. and accessible from a web browser
key performance indicators.” anytime, anywhere.
The Lift Manager extends the usual
This can be used to summarise all problem-solving capability of currently Huge step towards autonomous
basic operating parameters and available solutions by realising the buildings:
physical characteristics of a lift. This following benefits: ■
Shift from the current procedure
includes trips, door movements, of physical maintenance visits
number of floor trips, various vibrations, Securing and enhancing reliability and towards remote real-time
and also environmental conditions. safety for passengers: monitoring and predictive
■
Real-time lift operation status maintenance with an accuracy of
Anomaly Detection indicates any operational more than 80%.
“Lift Manager detects all essential anomalies or emergency situations.
lift conditions which exceed defined ■
Predicting degradation and failures,
thresholds of normal lift operation.” by using operational, environmental
and secondary data, secures
It is understood as lift events that passenger safety and operation
deviate from its normal operation availability by taking action before
parameters. Typical examples are potential lift failures occur.
mantrap, incorrect levelling of the ■
Optimising the availability of lift
cabin floor, door issues etc. operations.

6 Algorithm-based data analytics for lifts | TÜV SÜD

How does the solution work?
Implemented in an advanced IoT- FIGURE 1: GENERAL IOT-STRUCTURE OF LIFT MANAGER
environment, sensors are installed
on the lift to record the crucial daily
operation parameters in real-time.
Examples of sensor parameters
include three-dimensional
accelerations, environmental
conditions, and air pressure or
magnetic field alternations. These
parameters are later processed,
interrelated and integrated into
threshold-based patterns and IoT-Gateway
conditional networks. This upfront Sensor Kit
Analytics Process
process realises lean and efficient
data processing of complex to the corresponding cloud-service The role-specific user interface is
analysis. Another major benefit of via an IoT-gateway. The data is then designed using system-ergonomic
edge processing is significant data processed by means of rule-based and software-ergonomic principles,
transmission cost savings and a algorithms, advanced probabilistic which enables the performance of
reduction of related costs. models and AI-applications. Finally, user functions in a most efficient and
the results are made available to the comfortable manner.
The data is transferred over the air user via an interface or dashboard.

What makes our solution unique?

Current OEMs’ or IoT-companies’ happened during lift operation with FIGURE 2:
remote operation monitoring solutions what will happen in lift operation, by INTERDISCIPLINARY APPROACH
lack either brand independence using the collected empirical data and
for objectivity or the essential lift data analytics procedures. This is a Sensor kits for monitoring basic lift information
and predictive maintenance solution
domain know-how for reliability and completely new way of evaluating and
validity. The Lift Manager is a unique pre-interpreting data, which provides
combination of sensor and IoT- highly reliable, valid and accurate
expertise, as well as the independent results, as well as predictions after a
and integrative data analytics and lift very short “learning” time.
domain expertise of TÜV SÜD.
This unique deductive logic structure
The Lift Manager refrains from a solely is called a Lift Scenario Library (LSL),
inductive approach that exclusively as it uses probabilistic approaches
relies on data-driven probabilistic of the Fault Tree Analysis (FTA)
interpretations. For the Lift Manager, following EN 610251, and Event Tree
TÜV SÜD developed a 2-TIER-approach Analysis (ETA) following NUREG
that integrates this common inductive CR-28152 requirements. The LSL
approach with a deductively derived contains systematically related and
logic structure system. It achieves elaborated data about lift failures,
this by systematically combining the defects and deficiencies, which Provide algorithms based Data used for scheduling
lift domain expertise of what has TÜV SÜD has gathered for more on our analytic engines routine maintenance

TÜV SÜD | Algorithm-based data analytics for lifts 7

than 80 years as part of its testing, In detail, when a potentially abnormal 1. Green status:
inspection and certification business. behaviour is detected, the Lift Good for now. Issues can be
This logic structure system not only Manager’s predictive maintenance revisited at a later time
performs pre-channelling to interpret module prompts an alert to the 2. Orange status:
monitored lift events and process dashboard. This alert indicates when Some time until recommended
data (in terms of relevance and the related lift component or lift actions are required to be taken
significance) it also provides definitive subsystem is expected to fail, which 3. Red status:
maintenance recommendations and is known as “The Remaining Time The recommended actions should
prioritises them based on predictive to Downtime” (RTD). In addition, the be taken immediately
maintenance module results. This is a RTD has three priority categories
clear advantage that distinguishes the which indicate the urgency of the
Lift Manager from other solutions. recommended actions:

FIGURE 3: REMAINING TIME TO DOWNTIME (RTD) ALERTS

Good for now. Issues can be revisited at a

> 30 days
later time

Some time until recommended actions are

> 5 days
required to be taken
RTD

The recommended actions should be taken

≤ 5 days
immediately

8 Algorithm-based data analytics for lifts | TÜV SÜD

FIGURE 4: DEVELOPMENT USING TWO-TIER ANALYTICS PROCESS

Data Event
Tier 1: Data
interpretation identification
Added Value
Synchronised development & cross validation Root Causes
Recommended actions
Lift condition Event
Tier 2: Domain
interpretation identification

The LSL also includes information Over time, data generates new
about predicted or observed events: knowledge! The Lift Manager’s
algorithms are capable of learning The Lift Manager’s
1. LSL Root Causes and adapting from the ever-growing algorithms are capable
The LSL relates each observed database of sensor readings and
or predicted lift event to its root prediction validations. Validity of
of learning and adapting
causes or root cause scenarios. the Lift Manager’s observations and from the ever-growing
The deficient component or predictions are achieved from on-site database of sensor
subsystem can be repaired or validations and maintenance visits, readings and prediction
replaced before the failure event which are evaluated by lift experts
occurs and the safety or availability and inspectors. For accuracy, the data
validations.
of lift operation is restricted. analyst and data scientist toolbox is
2. LSL Recommended Actions enhanced, as it is based on sound
Each event and its set of root existing observation and prediction methods allow the Lift Manager to
causes and root cause scenarios algorithm sets. Advanced AI-methods, establish a highly reliable monitoring
is linked to a proven priority list such as pattern recognition and and prediction system in a short time
of recommended maintenance machine learning, continuously period of a few months.
actions, which enables the event to improve and extend accuracy and
be fixed effectively. future predictions. Altogether, these

FIGURE 5: ANALYTICS DEVELOPMENT & OPTIMISATION PROCESS

Development using two-tier Continuous onsite Continuous enhancement

analytics process verification & tailoring to lifts

Tier 1: Data Tier 2: Domain

Lift technicians feedback
Synchronised development & cross validation Corrective actions in
Data Lift condition the sequence of:
interpretation interpretation Continuous improvement
Most Likely
Event Event ↓
identification identification Likely
↓ Analytic engine

Added Value
Least Likely
Root Causes
To fine tune the likelihood of possible
Recommended actions
causes and recommended actions

TÜV SÜD | Algorithm-based data analytics for lifts 9

The Lift Manager solution at a glance
The Lift Manager solution is a
comprehensive combination of
IoT know-how and TÜV SÜD’s
independent and integrative
approach that fuses data analytics
and lift domain expertise. With
its elaborate data analytics
methodology and domain-specific
analytics engine, the Lift Manager
offers lift operators, OEMs, and
maintenance service providers:

■
A powerful and high-performing
integrated solution for condition
monitoring and predictive (TIC) expert organisation, which has ■
More than 80 years of experience
maintenance. experience with every lift portfolio. in lift inspections are crystallised in
■
An independent solution from a ■
The technical know-how of the unique TÜV SÜD LSL.
globally acknowledged third-party, integrating software, sensors and
Testing, Inspection and Certification cloud services.

About TÜV SÜD

As a classic domain expertise We rely on more than 100 years of and are renowned for independent
provider in the field of TIC services, knowledge and a worldwide network inspection, acceptances and
TÜV SÜD has always been closely of over 750 elevator experts. We evaluation of elevator components,
linked to its customers and partners. conduct 300,000 inspections annually as well as complete elevator systems.

10 Algorithm-based data analytics for lifts | TÜV SÜD

FOOTNOTES
[1] BS EN 61025. Fault Tree Analysis (FTA)
[2] NUREG/CR-2815. Probabilistic Safety Analysis Guide

COPYRIGHT NOTICE
The information contained in this document represents the current view of TÜV SÜD on the issues discussed as of the date of publication. Because TÜV SÜD must respond to changing
market conditions, it should not be interpreted to be a commitment on the part of TÜV SÜD, and TÜV SÜD cannot guarantee the accuracy of any information presented after the date of
publication. This White Paper is for informational purposes only. TÜV SÜD makes no warranties, express, implied or statutory, as to the information in this document. Complying with all
applicable copyright laws is the responsibility of the user. Without limiting the rights under copyright, no part of this document may be reproduced, stored in or introduced into a retrieval
system, or transmitted in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), or for any purpose, without the express written permission of TÜV
SÜD. TÜV SÜD may have patents, patent applications, trademarks, copyrights, or other intellectual property rights covering subject matter in this document. Except as expressly provided
in any written license agreement from TÜV SÜD, the furnishing of this document does not give you any license to these patents, trademarks, copyrights, or other intellectual property.
ANY REPRODUCTION, ADAPTATION OR TRANSLATION OF THIS DOCUMENT WITHOUT PRIOR WRITTEN PERMISSION IS PROHIBITED, EXCEPT AS ALLOWED UNDER THE COPYRIGHT
LAWS. © TÜV SÜD Group – 2021 – All rights reserved - TÜV SÜD is a registered trademark of TÜV SÜD Group

DISCLAIMER
All reasonable measures have been taken to ensure the quality, reliability, and accuracy of the information in the content. However, TÜV SÜD is not responsible for the third-party content
contained in this publication. TÜV SÜD makes no warranties or representations, expressed or implied, as to the accuracy or completeness of information contained in this publication.
This publication is intended to provide general information on a particular subject or subjects and is not an exhaustive treatment of such subject(s). Accordingly, the information in this
publication is not intended to constitute consulting or professional advice or services. If you are seeking advice on any matters relating to information in this publication, you should –
where appropriate – contact us directly with your specific query or seek advice from qualified professional people. The information contained in this publication may not be copied, quoted,
or referred to in any other publication or materials without the prior written consent of TÜV SÜD. All rights reserved
© 2021 TÜV SÜD.

TÜV SÜD | Algorithm-based data analytics for lifts 11

Contact us for more information
www.tuvsud.com/liftmanager
[email protected]

Add value. Inspire trust.

TÜV SÜD is a trusted partner of choice for safety, security and sustainability solutions. It specialises in testing,
certification, auditing and advisory services. Since 1866, the company has remained committed to its purpose of
enabling progress by protecting people, the environment and assets from technology-related risks. Through more than
24,500 employees across over 1,000 locations, it adds value to customers and partners by enabling market access
and managing risks. By anticipating technological developments and facilitating change, TÜV SÜD inspires trust in a
physical and digital world to create a safer and more sustainable future.

2021 © TÜV SÜD AG | MKG/RE/40.2/en/SG

TÜV SÜD PSB

15 International Business Park
Singapore 609937
+65 6973 6444
www.tuvsud.com