Maintenance 4.0 is the application of Industry 4.0 to operations and maintenance (O&M) activities. The goal is simple: To maximize production uptime by eliminating unplanned, reactive maintenance. Let’s look at a simplistic depiction of common O&M work streams.
Figure 1 shows a graph depicting the activities that occur
after an industrial asset unexpectedly fails.
Figure 1: O&M work streams in Industry 3.0 vs Industry
4.0
Once the failure event occurs and is
reported, a series of activities occurs. First, repair crews are assigned and
then travel to the worksite where they receive repair instructions. Parts must
be ordered and transported to the site.
Typically, root cause analysis (RCA)
is performed and valuable time expended on identifying it. Working under
pressure to resume production, work crews engage in trial and error activities
to identify the cause of the failure. After repairs and an inspection,
production resumes.
Maintenance 4.0 brings artificial
intelligence (AI) and machine learning (ML) to the production line. Instead of
waiting for the equipment to fail, sophisticated algorithms are applied to big
data from embedded sensors in the equipment. The algorithms are trained to identify
correlated patterns of anomalous machine behavior and warn of evolving machine
failure.
Figure 2: Core elements of Maintenance 4.0 (Source: Presenso).
Figure 3: Cost comparison for storage, bandwidth and computing from 1991 to 2019 (Source: Deloitte Consulting)
Within Maintenance 4.0, AI-driven
industrial analytics is the game changer.
Until recently, machine learning was a study confined
mostly to academia. A confluence of multiple factors has lowered the cost of data
transportation, bandwidth, storage and analysis. For example, data storage has
fallen from five hundred and sixty-nine dollars per gigabyte in the early 1990s
to less than one cent today.
Figure 4: Detection of evolving
failures using machine learning (Source: Presenso)
Figure 1-5: Reactive maintenance processes (Source: Presenso)
As a result of the cost decline, machine learning can now be
applied to vast amounts of sensor-generated big data that can be analyzed in real
time.
The first component of Maintenance 4.0 is that while the
failure is evolving, repairs can be scheduled and parts ordered. Tracing the
failure to the original root cause eliminates guesswork and trial and error.
With Maintenance 4.0, machine uptime can be maintained while all
non-repair activities are executed.
The second component of Maintenance 4.0 is the adoption of a
computerized maintenance management system (CMMS) and automated workflows.
Although a CMMS is not new, until now, its implementation has not been
considered of strategic importance.
The third element of Maintenance 4.0 is the use of robotics
and drones for inspections and repair activities.
In 2018, research was conducted to
gain insight into industrial plants’ plans for the adoption of Maintenance 4.0.
Figure 6 shows the results of that study.
Figure 6: Survey results regarding industrial plants’ plans
for Maintenance 4.0 (Source: Emory University and Presenso)
Source: https://industrial-ai.skf.com/the-maintenance-4-0-implementation-handbook-2/
Comments
Post a Comment