Munich. Building a car generates vast amounts of
information via a value chain. The BMW Group uses a Smart Data
Analytics digitalisation cluster to analyse this information selectively and
raise a prolongation system. Results from intelligent information analysis
make an effective grant towards improving peculiarity in all areas
of prolongation and logistics.
Data-driven improvements to processes and systems assistance revoke lead
times and reduce costs. New solutions are being grown not usually at
headquarters, though during many opposite points in a international
prolongation network. In this way, a BMW Group is means to take
advantage of a far-reaching operation of innovations that open adult additional
options for even some-more stretchable production. The association uses an
access-protected intranet-of-things height to couple a large
apportion of sensor and routine information from prolongation and logistics
fast and easily. Smart Data Analytics therefore offers completely
new opportunities that extend distant over prior analysis
possibilities. The speed with that new solutions can be implemented
is augmenting significantly. At a same time, new IoT sensors,
total with cloud and vast information technologies, are shortening the
technical complexity and doing costs involved
Christian Patron, conduct of Innovation and Digitalisation in Production
System: “Smart Data Analytics is environment new standards for our
prolongation system. By mixing a knowledge of a staff with new
possibilities for fit estimate of vast information volumes, we are
means to emanate accurate forecasts and proactively optimise processes.
This speeds adult continual alleviation of a prolongation complement in line
with a simple beliefs of gaunt production.
Numerous use cases implemented in several prolongation sections
denote a advantages of Smart Data Analytics.
Laser-marked physique parts: fine-tuning for presses; physique parts
traceable during all times
Steel coils adult to 40 tonnes in
weight and about 3 kilometres prolonged are cut into blanks in the
press emporium and afterwards shaped into physique parts. However, piece thickness,
strength, aspect hardness and grade of oiling are not uniform
via a coil. Deviations from aim can lead to cracks in body
collection that are subjected to sold highlight during a forming
process. This is where a Smart Data Analytics concentration during BMW
Group Plant Regensburg comes in. A laser is used to symbol any blank
with a multi-digit code, that serves as a possess ID. Going forward,
this ID will concede a presses to be fine-tuned to accommodate the
characteristics of a blank. If needed, a ID might enclose a control
command, that triggers additional oiling of a vacant in a press
before forming, for example.
This transparent imprinting enables a vacant to be identified during any time.
Each physique partial is reserved information that stays available
via all successive prolongation steps. Since a vacant stays in
a prolongation line for marking, a ID is reserved though any cycle
downtime. The ID is designed so that it stays manifest via car
physique construction. BMW Group formulation specialists already take
advantage of a traceability of all collection for serve optimisation
involving additional algorithms. For example, holding into comment the
characteristics totalled for any sold physique part, a gap
measure of a finished physique can be serve optimised, or a paint
concentration improved matched to a aspect of that sold body.
Fine-tuning of press parameters according to a properties of the
vacant is already carrying a vital impact: The array of scrapped parts
is significantly lower, with improved utilization of a curl material.
The complement downtime compulsory for blunder research is also reduced.
Predictive upkeep for physique emporium robots, welding collection and
Smart Data Analytics applications offer
generally high intensity for augmenting a accessibility of
prolongation apparatus and machines in highly-automated areas of
manufacturing. Maximum correctness in presaging any risk of breakdown
mostly helps equivocate random complement downtime. Based on a forecast,
upkeep staff can devise a targeted upkeep involvement to
extent downtimes to an extensive minimum. This supposed predictive
upkeep is enabled by intelligent research of vast quantities of
genuine production, sensor and routine data: Targeted research of this
information creates it probable to establish a ideal time to replace
wearing collection used in production. If a change is done too late,
there is a risk of prolongation stoppage; done too early, valuable
resources are wasted. Without a applicable information on that to bottom this
decision, a quite surety upkeep of a past was conducted
though meaningful a tangible state of wear. This routine required
permitting reserve margins for a timing of a changeover, though could
not detect astonishing breakdowns.
Data-based solutions for predictive upkeep are used during various
stages in car physique prolongation to envision rigging and stop wear in
robots. Sensors in welding tongs vigilance forward of time when defects or
peculiarity problems are expected to occur. Widescale sensor monitoring also
improves a trustworthiness of a electrical drives used in a accumulation of
systems, including rises and turntables. Robots and control technology
are propitious with a required sensors from a start. Maintenance
staff analyse a information and afterwards pull a right conclusions. Recent
evaluations of predictive upkeep clearly denote a benefits
for arguable operations.
Online routine controls: Even some-more fast processes guarantee
The BMW Group perceived a Prix de la Technique 2017 during the
prestigious Surcar Congress in Cannes for a judgment for
extensive paint emporium digitalisation during a company’s new plant in
San Luis Potosí, that will start array prolongation in 2019. BMW Group
paint shops already use sensors for ongoing monitoring of automated
prolongation processes. Intelligently networked systems raise the
fortitude of routine sequences, capacitate predictive upkeep and
safeguard a top peculiarity for a customers. Online routine control
combines a strengths of algorithm-based research of vast data
volumes with worker experience: As a result, humans can concentration more
on their purpose as architects of a prolongation process, given real
prolongation information is sorted and optimally pre-structured for them. Error
intensity can be rescued in time and redo avoided.
In May 2017, a BMW Group began regulating fully-automated quality
control for a initial time during a Munich plant, with robots scanning
a whole outdoor car surface. The complement is able of detecting
errors a tellurian eye can't perceive. The information performed in this way
also provides profitable feedback on a pointing of upstream painting
processes – permitting continual optimisation and timely identification
of forsake potential.
Fastener information analysis: More arguable blunder prevention
advantages thousands of bolted connections
connectors are elemental to car production, given every
car contains several hundreds of them. The BMW Group monitors and
analyses all bolted connectors that are applicable to a reserve of the
vehicle. Basically, bolted connectors that do not, or usually partly,
accommodate a preferred specifications might need rework. As partial of its
surety peculiarity strategy, a BMW Group has grown algorithms
that have been analysing bolted connectors in some-more than 3,200
public systems during all car plants given Jul 2017. Recording and
research of bolting routine curves yield accurate feedback on the
peculiarity of bolted connections. The programme can recognize different
forms of blunder and uncover probable sources of blunder in a
cause-and-effect diagram. The BMW Group uses this information to train
and validate employees for surety peculiarity work – after all, a
mistake that is not done does not need correcting. A tutor during a
mobile training hire or directly during a workplace can also provide
tips on blunder avoidance.
Analysis of bolting routine curves also provides critical insights
for systematic monitoring of bolting systems and parameters, such as
tightening torque. When implemented quickly, these commentary emanate a
sealed loop of continual improvement.
In many cases, quite primer research of bolting routine curves would
usually outcome in a anticipating of “acceptable” or “not
acceptable”, though identifying a means of errors or
highlighting intensity for improvement.
Predictive upkeep for materials doing in
The BMW Group prolongation complement is
characterised by a top grade of flexibility: The company
produces an generally far-reaching operation of models and variants on its
public lines, though is but rival – as arguable by
eccentric benchmarks. On a public line, a arguable supply of
materials is quite important. A relapse during any indicate could
means a whole prolongation area to grub to a halt.
In assembly, many circuit systems are now versed with a large
array of sensors that guard several factors – especially
temperature, quivering and electrical power. These sensors are
cost-effective adequate to concede them to be widely used. Data from these
sensor kits and other routine information is streamed live to a BMW
internet-of-things platform, where it is visualised and analysed in
real-time. If a information detects a trend indicating flaw or
patterns from prior breakdowns, a height notifies maintenance
staff. Staff can afterwards confirm either a hanger should be private for
maintenance. In this way, it is probable to safeguard long-term, reliable
operation of a circuit complement over a array of years.
Every notation saved means another vehicle
Christian Patron: “In automotive production, every
second counts: If a partial isn’t accessible on time or a complement fails,
a prolongation routine is behind and it disrupts a value chain.
Intelligent use of prolongation information ensures a fast and efficient
process. We see extensive intensity in Smart Data Analytics for
incorporating feedback from a business into growth and
prolongation even faster.”
At a BMW Group, record supports people. Now and in a future:
Here we can find a YouTube Playlist.