Geen industrie 4.0 zonder onderhoud 4.0
Wim Vancauwenberghe - Directeur BEMAS (BELGIAN MAINTENANCE ASSOCIATION)
de cruciale rol van onderhoud in Industrie 4.0
onderhoud 4.0: wat, hoe en waarom?
enkele praktische voorbeelden en cases van onderhoud 4.0
2. Vision
Maintenance & Asset Management play a key role in achieving a sustainable
industrial activity in Europe. Managers nowadays realise that an efficient
maintenance and asset utilisation leads to increased benefits. A continuous
improvement of asset utilisation and the reduction of (production) costs,
results in added value for all stakeholders. As a result, maintenance and asset
management are to be considered as key drivers to create competitive
advantage and sustainable operation.
Mission
Help (Belgian) asset owners to increase their competence level by sharing
knowledge on maintenance, production reliability & asset management, and
by creating a larger awareness and appreciation for maintenance activities
and the responsible maintenance & asset managers.
About
3. BEMAS – Belgian Maintenance Association vzw-asbl
www.BEMAS.orgMember of
5. Content
• Introductie
• De cruciale rol van onderhoud in Industrie 4.0
• Onderhoud 4.0: wat, hoe en waarom nu?
• Enkele voorbeelden en cases van onderhoud 4.0
6. Wat is industrie 4.0 ?
Bron: Industry 4.0 - de huidige stand van zaken
in de Vlaamse industrie – maart 2017
8. Wat is industrie 4.0 ?
Industry 4.0 is a name for the current trend of automation and data exchange in
manufacturing technologies. It includes cyber-physical systems, the Internet of
things, cloud computing and cognitive computing.
Industry 4.0 creates what has been called a "smart factory". Within the modular
structured smart factories, cyber-physical systems monitor physical processes,
create a virtual copy of the physical world and make decentralized decisions.
Over the Internet of Things, cyber-physical systems communicate and cooperate
with each other and with humans in real time, and via the Internet of Services,
both internal and cross-organizational services are offered and used by
participants of the value chain
Bron: https://en.wikipedia.org/wiki/Industry_4.0=> Maintenance 4.0
10. Content
• Introductie
• De cruciale rol van onderhoud in Industrie 4.0
• Wat is Onderhoud 4.0 en waarom nu?
• Enkele voorbeelden en cases van onderhoud 4.0
11. Is Maintenance important?
• In Europe, about 6 million people
work in technical maintenance
• Every year about 450 Billion Euro is
spent on maintenance of industrial
technological assets, with an
estimated reinvestment value of
10.000 Billion euro. A big part of
these industrial assets are based in
NWE.
11
19. INVEST in Maintenance 4.0
? % extra meer
? % extra minder
? % minder correctief en dus meer preventief
? % minder periodiek tov huidige best performers
? % extra minder tov huidige best performers
20. Content
• Introductie
• De cruciale rol van onderhoud in Industrie 4.0
• Wat is Onderhoud 4.0 en waarom nu?
• Enkele voorbeelden en cases van onderhoud 4.0
21. Waarom nu ?
• Nieuwe kijk op productie
• Lean -> Agile
• Mass Production -> Mass Customisation
• Generatiewissel bij engineering – maintenance – operations
• Investeringscyclus
22. Remaining lifetime of Plants
22 Sample size: 82
Plants = process industry +manufacturing+food, beverage & pharma
44 % of the ARV reaches end of life within 10 years !
29. Platform positioning is important
Source: Forrester Research
Also Siemens (Mindsphere), ABB (Ability), Bosch,
Rockwell, SKF, … are developing/ offering (sub)
platforms
30. IoT is a SCALE BASED BUSINESS /
PLATFORM
PLATFORM ECONOMICS
• Network Effects
— The more customers join, the more valuable the
platform (subsidies possible)
• Demand side economies of scale
— Fixed cost amortized over large number
• Winner takes all markets
— Very high market shares top players; Difficult for most
competitors
THIS IS WHAT OEMs and TECHs ARE TRYING TO DO
31. Content
• Introductie
• De cruciale rol van onderhoud in Industrie 4.0
• Wat is Onderhoud 4.0 en hoe begin je er aan?
• Enkele voorbeelden en cases van onderhoud 4.0
32. Today‘s Predictive Maintenance
• Predictive Maintenance provides significant value, while
reducing outcome and performance risks
• Predictive Maintenance
— Prevents unexpected equipment failure
— Improves uptime, availability and reliability of plant
— Allows scheduling of corrective interventions and
reduces maintenance costs as most interventions
performed only when warranted and planned far more
cost effective than unplanned maintenance
• Includes
— Failure prediction (anomalies) – based on expert
opinions
— Failure detection and failure type classification
(diagnosis, root cause analysis)
— Recommendations for mitigation or maintenance
actions before and after failure
33. LOW FIXED / HIGH VARIABLE COSTSBut PdM could NOT scale
• Traditionally based on conventional
Condition Monitoring and expert (human)
analysis to provide diagnosis: High
VARIABLE cost, expert dependent, error
prone limited scope (critical expensive
assets), only some customers
• Numerous efforts made to automate
(computerize) prediction / analysis over past
15 years or so, but limited applicability (cost,
generalization) – deductive approach,
difficult model building
Source:
34. HIGH FIXED / LOW VARIABLE COSTSIIoT means PdM CAN
NOW scale
• Now IIoT (sensors, big data, connectivity,
cloud) + analytics enable an automated,
inductive approach. Machine learning
algorithms can potentially automate
processes, reduce/eliminate VARIABLE
costs, improve accuracy (reduce errors),
extend time horizons and provide new
insights
• Make no mistake: Automation means
significantly removing (net) human (expert)
intervention
Source:
35. Technical Advances As-a-Service
Break
& Fix
Today
Predict
& Prevent
Sensors,
Instruments,
Algorithms (FFT)
Condition
Monitoring
Predictive
Maintenance
RBM / RCM
Sensors,
Connectivity
(IoT), Analytics,
Algorithms,
Apps
Prognostics,
RUL, CPS
Intelligent
Maintenance
Systems,
Industrie 4.0
Condition and Remote Monitoring created an additional market worth > 5B$, but reduced maintenance
expenditure by reducing unnecessary interventions (deadweight costs) and improving productivity
36. PdM predicts timing, probability of failure
Performance/
Condition
Operating Time
Normal operating band
Anomaly detection threshold
Predictive
Diagnostics
Prognostics
RUL probability distribution
Diagnosis threshold
RUL curve
Functional Failure threshold
Source:
37. Analytics based PdM are a developing field
(no standard solutions yet)
Physical Models
• Mathematical/Accuracy?
• Domain Expertise
• Verification (data)
• Asset/Component specific
• High Cost
• Critical/Expensive assets, e.g. turbines
Knowledge/ Expert Models
• Rules based
• Non-generic
• Cannot handle situations not covered by rules
• Risk of combinatorial explosion from too many
rules
Fuzzy Logic Models
• Vagueness and non-linearity (inputoutput
probability distributions)
38. Analytics based PdM are a developing field
(no standard solutions yet)
Big Data Driven Models
(Induction of generalized models from empirical data – Machine Learning)
• Supervised learning: Data collected from past, filtered, labeled according to expected result
• Unsupervised learning: Historic data not labeled, Scope: Discover groups of similar
examples in historic data (leaves machine to find and structure its input)
• Reinforcement learning: Interaction with dynamic environment, feedback-driven, rewards
and penalties
Regression: Predict the Remaining Useful Life (RUL), or Time to Failure (TTF).
Binary classification: Predict if an asset will fail within certain time frame (e.g. days).
Multi-class classification: Predict if an asset will fail in different time windows: E.g., fails in
window [1, w0] days; fails in the window [w0+1,w1] days; not fail within w1 days
BASIC MODELING SOLUTIONS
39. Yes
Decision tree for predictive maintenance
Reinforcement
Learning
Expert Model
Fuzzy Logic Model
Unsupervised
Learning
Supervised
Learning
Vague, imprecise,
noisy or missing
inputs?
Historical data
available?
Physical Model
Knowledge based
Model
Data Labeling
feasible?
Data driven
Model (ML)
Capability /
Resources for
physical model?
Data from real
world systems
available?
No
Yes
Yes
No
No
No
Yes
46. Content
• Introductie
• De cruciale rol van onderhoud in Industrie 4.0
• Onderhoud 4.0: wat, hoe en waarom nu?
• Enkele voorbeelden en cases van onderhoud 4.0
56. Robotisation - Drones
Unmanned flying drones are to be used by a
low-fare airline to inspect its fleet of Airbus
aircraft.
Budget carrier easyJet hopes to introduce the
drones as early as next year following trials in
the next few months.
The drones will be programmed to scan and
assess the carrier's fleet of Airbus A319 and
A320 planes, reporting back to engineers on
any damage which may require further
inspection or maintenance work.
The airline is working with the Coptercraft
and Measurement Solutions companies as
well as Bristol Robotics Laboratory on
modifying existing technology so it can bring
in the drones. Ian Davies, head of engineering
at easyJet, said: "Drone technology could be
used extremely effectively to help us perform
aircraft checks.
http://www.westerndailypress.co.uk/Airline-easyJet-ditches-pilots-
maintenance/story-21071854-detail/story.html#ixzz3UfSPIHV1