updated. If planning is only trued up every
one, three or six months, the loss of efficiency
from slow planning can be significant and often difficult to recoup.
We believe that thinking about supply chain
optimization from a digital twin perspective, covering all the components that can
actually impact success or failure of a given
supply chain strategy--such as product quality and performance, supply reliability, manufacturing capacity and lead time, inventory
availability, and logistics effectiveness, and
demand changes--is essential to success. Enabled by a modular, component-based modeling architecture, we have been able to compose digital twins to answer specific questions
about the performance of individual elements
of the supply chain and help drive performance to a new level. And the twins are kept
up to date with the latest information from
the real world and learnings from the digital
world using continuously improved predictive models that learn from new data from the
physical world. This level of real-time concurrency is critical and marks a stark differentiation from how much traditional supply chain
planning is done today.
The digital twin for manufacturing and
supply chain optimization is essentially the
digital version of the physical world, based on
historical, current, and predicted future information as to the state of materials, parts, component, assemblies, finished goods, logistics,
deliveries, and so on. Thinking about supply
chain optimization as a complex network of
twins in a multiverse of potential realities allows us to take advantage of parallel processing versus sequential reasoning.
We at GE are developing the software and
computational systems to support this view,
and a new approach to solving our seemingly
intractable problems is emerging. By creat-
ing and leveraging hundreds of thousands
of digital twins across GE’s global supply
chain operations, we are able to drive new ef-
ficiency into our operations and achieve sig-
nificant improvements in systems lead time,
capacity, inventory and cost.
At the same time, by developing digital
twin modular apps for monitoring, optimization, planning, and execution across
the supply chain, we are enabling a flexible
approach to implementing industrial out-comes-based solutions, focused on the biggest problems and/or the highest returns
on investment. The systems footprint of a
global supply chain can be very large, spanning hundreds or thousands of suppliers and
customers, logistics systems, warehouse systems, and production systems. Monolithic
solutions to managing and optimizing these
systems, such as enterprise resource planning systems that focus on a single company
or sub-business, are unlikely to be sufficient,
as the implementation time is typically too
long and expensive and the solutions too
constrained to be able to rapidly exploit
emerging opportunities across the extended
and increasingly complex supply chain systems of systems that span multiple companies and business units.
We believe that modular applications that
take advantage of underlying digital twin
technologies are core to successful supply
chain optimization strategies. Each application focuses on a specific opportunity – such
as lead time improvement, inventory reduction, sourcing cost reduction, etc. – as part
of a bigger portfolio of applications that all
exploit digital twin technology. This not only
ensures coherency across the app portfolio,
but also enables concurrency of learning and
optimization based on the latest information
available from the physical world.
Digital twin ecosystems modeling takes
the application approach to a systems level.
................... ................... ................... ................... ................... ................... ...................
we are able
to drive new
into our op-