Skip to main content

Re-building fragile global supply chains with graphs

(Image credit: Image source: Shutterstock/KAMONRAT)

As the coronavirus sweeps the globe, closing businesses and forcing knowledge workers to work from home, we are all reminded of how important supply chains are to every part of our lives — and how easily they falter.

As borders shut and panic buying of items and shortages of vital medical equipment ensue, many are wondering if the root cause might be the opacity and the rigidity of supply chains that easily collapse under strain. While no-one could have predicted the scale and the speed at which the pandemic unfolded, could we have been better prepared? It’s a problem summed up by The World Economic Forum, which warns that, “Governments, businesses and individual consumers suddenly struggled to procure basic products and materials, and were forced to confront the fragility of the modern supply chain. The urgent need to design smarter, stronger and more diverse supply chains has been one of the main lessons of this crisis.”

We need robust and adaptive supply chains. We’ll likely see a roll-back from over-dependence on single sources for goods, but modern capitalism relies upon global markets. As a result, manufacturers, distributors and logistics companies all need a more agile way of dealing with the vast amount of intertwined data and regulations involved with delivering items around the world.

It is important that we step back and consider what we can learn from Covid-19. Hopefully, businesses can use that knowledge to build stronger, scalable and more flexible supply chains that will help us kick start economic activity. To achieve this they will need a better understanding of the data flowing in and out of their supply chains, so they can gain a real-time insight for smart decision-making. At the same time, brands may need to win back consumer and customer confidence, and in some cases, loyalty. All of this needs to happen while ensuring we deliver products that meet international standards and regulations and maintain our own standards of sustainability and social responsibility, as quickly as possible.

The ripple effects of the pandemic are putting companies at risk of delivering products that are below par or don’t meet regulations. Sub-standard components may be hastily ushered into the supply chain without being scrutinised and could place manufacturers’ entire operations in a perilous position. This poses additional risk in closely-regulated industries such as pharmaceuticals or medical device makers, where suppliers must be able to identify and locate an individual item or batch at any given time.

The more complex a supply chain, the more vulnerable it is, maintains supply chain technology experts Transparency-One. Its CTO, Frédéric Daniel, for instance, believes that supply chains that incorporate multiple tiers, are heavily globalised and/or involve a number of components or stages of transformation are inherently more complicated and at greater risk of being broadsided by a crisis like Covid-19.

However, Daniel also contends that a supply chain’s vulnerability is partly down to how prepared you are to deal with a crisis. Companies that have 360-degree visibility of their supply chains and supplier ecosystem know how production will be impacted. They will quickly realise that they need to look for alternative sources if there is a shortage of components, for example, or if ports are locked down. Those who are not prepared for this, or indeed the next black swan event, will find it almost impossible to mitigate supply shock and manage associated demand volatility. 

Unravelling supply chain complexity

In an ideal scenario supply chains would be a linear chain of single suppliers. Unfortunately, real life is much more complicated and supply chains are very hard to consider holistically. Many enterprises still have their data stored in silos, meaning they only have a partial view of what is going on in their supply chains. And even if the data is stored in a single relational database, understanding the connections between products on a production line or in pallets waiting to be shipped is next to impossible.

As we move to Industry 4.0 and the Internet of Things where data and processes are interdependent, there is an increased ability to gain data driven insights and at the same time a rise in overall complexity. Relational database technology, which stores data in rows and columns, is poorly-equipped for identifying relationships within datasets, but these connections are imperative for identifying a product’s whereabouts as well as monitoring, analysing and visualising the supply chain. These connections also need to be quick and easy to search and adaptable to the size of the supply chain.

Making traditional databases perform multidimensional tasks in real time is also very difficult, with performance degradation as the total dataset size grows. CIOs need a scalable, agile way of managing thousands of different product lines, produced across multiple sites, which are sold into hundreds of diverse markets. The number of unique serial codes alone can run into the billions very quickly.

The good news is that there is a solution, in the shape of graph database technology. This solution can record and connect the dots between complex data interdependencies, the concept being that when you track a product or component you create a hierarchy of data and store how it’s all related. Performance is maintained, even with vast quantities of data; and if you scan the code on a particular pallet, it can automatically recall not only all of its contents but also the context such as which ports they shipped through, when they were manufactured, and even the relationships between manufacturers.

Instead of utilising relational tables, graphs use structures better at analysing interconnections in data. Graph data models are flexible and do not need to be hard coded, making a graph database practically impossible to beat at analysing the relationships between a large number of data points. Such a connected relationship-centric approach allows businesses to better manage, read and visualise the data in lengthy and complex supply chains.

With greater visibility into their supply chains, manufacturers can drill down to gain an accurate, trackable picture of products and their whereabouts. It can also ensure that its practices are sustainable. Using a graph database, manufacturers can typically demonstrate 100 times faster query response speeds than that enabled by SQL RDBMS software. This sort of response time is critical during the present crisis and will be crucial going forward in a highly digitised, increasingly competitive world.

A smart supply chain strategy

This isn’t just a matter of digitising supply chains, it is about doing so with the most appropriate tools and technology that provide insight into complicated interconnected supply chains. It is also about being equipped with the best technology to adequately tackle our current reality — complex, interconnected supply chains. Finally, it’s about delivering the transparency and traceability required to enable manufacturers to rapidly identify risk and respond to disruption.

Covid-19 won’t last forever. In the coming ‘new normal’, it will be essential to put the right technology in place to provide insights from the data we already possess to run our businesses with the agility and flexibility needed to survive and thrive. Graph database technology could be a real enabler here, providing a collaborative platform where gargantuan amounts of connected data can be handled at scale.

This pandemic will not be the last crisis that global manufacturing faces. Having a smart supply chain strategy in place is an urgent requirement to combat future disruption — and graphs have an essential role to play.

Amy Hodler, Director, Analytics and AI Program, Neo4j