Page 3 - Logistics News April 2018
P. 3
Thought Leadership
A look into the future:
The self-learning supply chain
When ‘deep learning’ AI is incorporated into supply chain systems, they will be able to
analyse past supply chain failures in order to prevent new ones.
By Fred Baumann, Supply Chain Quarterly
THE SELF-LEARNING supply chain marks the move up the maturity curve, their reliance
next major frontier of supply chain innovation. on manual capabilities will be replaced with
It’s a futuristic vision of a world in which autonomous capabilities, providing them with
supply chain systems, infused with artificial significant efficiency gains and cost savings.
intelligence (AI), can analyse existing supply Most companies today are in the first
chain strategies and data to learn what stage of digital supply chain maturity – the
factors lead to supply chain failures. These visibility phase. Currently, there is a huge
AI-driven systems then use this knowledge focus on end-to-end supply chain visibility to
to predict future supply chain problems help companies better manage constraints.
and proactively prescribe or autonomously At this maturity stage, visibility is often
execute resolutions. While there is still a way enabled by various system integrations such
to go before the self-learning supply chain is as connecting enterprise resource planning
a reality, recent advancements in AI suggest it (ERP) systems with best-of-breed solutions
is no longer ‘blue-sky thinking’. and customer systems. This type of system
The self-learning supply chain of the integration enables a business to gain an end-
future marries the benefits of AI with the to-end view of how product flows through its
digital technologies that many companies supply chain.
have already started incorporating into The next stage of digital supply chain
their supply chain disciplines. This digital maturity is predictive analytics. This phase
supply chain transformation is being fueled leverages predictive analytic algorithms,
by several technology advancements: enabled by big data such as Internet of
Physical ‘things’ incorporating computer things (IoT) sensor data, SNEW data and
technology, readily available big data such others to predict where supply chain issues
as social media, news, events and weather may arise in the future. Predictive analytics,
(SNEW) and computer systems and software for instance, can be used to analyse real-
becoming more intelligent. These digital time data like weather forecasts and
technologies are transforming the very nature port congestion to predict the impact on
of the supply chain, which was once built freighters in route and determine which
for volume and scale, into an agile, digitally shipments will be late – even before the
connected framework that leverages a single captain may know.
set of physical assets to support multiple The prescriptive supply chain, enabled
virtual supply chains. These virtual supply by supervised machine learning, is the next
chains replace the traditional fixed linear stage of digital supply chain maturity. In this
supply chains of the past by providing new stage, intelligent systems will be able to move
flow options that enable accelerated order beyond predicting potential supply chain
fulfilment based on near real-time awareness issues to prescribing the course of action to
of assets and inventory. take to resolve the issue. This technology
is already being incorporated into best-of-
The path toward the digital supply chain breed offerings, where prescriptive analytics
We predict that the path toward digital are used to learn from planners’ historical
supply chain maturity will occur in four actions. For a shipment that’s predicted to be
stages: Visibility, predictive analytics, the late, for instance, the solution could provide
prescriptive supply chain and, ultimately, several resolution options (such as swap
the self-learning supply chain. As companies demand from another resource or purchase
2 April 2018 | Logistics News
