Whether it’s the escalating tariffs between the U.S. and China, the status of so-called ‘new NAFTA’ or what’s happening (or not) with Brexit in Europe, what’s clear in the industry today is that there are unprecedented uncertainty and risk in global supply chains. This ever-changing political landscape makes it difficult for companies to know which direction to take.
Supply chains are inherently complex, inter-connected systems with many interdependencies. A decision which offsets a tariff increase or otherwise improves performance in one area can drive disproportionate cost or waste in another. And, not only are the optimal trade-offs complex, but they can often also be counterintuitive.
So, how can organisations maintain high-performing, cost-effective supply chains when the sands are constantly shifting?
One of the tried and tested approaches to managing uncertainty and risk in the supply chain is to have a framework for scenario planning. This is not a new concept – companies have been scenario planning for more than 30 years – identifying critical events before they occur and using this knowledge to determine effective alternatives.
Thanks to globalisation, what has changed is the degree of complexity and, with that, the sheer range of possibilities and eventualities. Until relatively recently, the time and effort required to explore any more than the most likely scenarios has been prohibitive.
What has also changed is the unprecedented amount of data that’s available both from within and outside an enterprise.
It’s not enough to model your physical network from a location, shipment or order perspective. Making the right decisions, considering all trade-offs and constraints, requires both a big picture view of the end to end supply chain and a granular understanding of product flows, assets, customers and the mechanisms by which their requirements are met.
To achieve this, we see more and more leading organisations turning to technology to build ‘digital twins’ of their real-world supply chain, providing them with a risk-free environment in which to quickly ask and answer unlimited ‘what-if’ questions, from the expected to the unlikely. This supports evidence-based decision making and allows them to embark on both supply chain optimisation and strategic transformation programmes with confidence that the changes they make will deliver the desired and expected outcomes.
The availability of data makes this possible, however, harnessing that data still requires some heavy lifting up front.
Some 20 years ago, 80-85% of the time taken to build a supply chain model was spent on data collection, cleansing, harmonisation and blending to build a base case scenario. Although we are at a different time and age now, the effort required to build a solid data foundation should not be underestimated.
The difficulty is that the data required to build a digital twin of the supply chain resides in a patchwork of systems – transport management, warehouse management, demand planning, sourcing, inventory management – the list goes on, each designed to support a single supply chain function. There is no single supply chain system of record in any enterprise today, nor any appetite for creating one.
However, you can create a single system of reference for the supply chain, where you are able to understand the logical dependencies and model and manage the different behaviours of the different elements of your supply chain. That requires an integrated infrastructure that lets you tap into multiple Enterprise Resource Planning (ERP) and supply chain planning and execution systems, as well as external systems. And, although there are some tools and best practices to help you do that, there is no short-cut.
However, once you’ve done it – once you have your system of reference established with the right integration touchpoints across your systems of record – then you can start formulating and playing out all the different scenarios and using that framework to make decisions about the future.
To coin a phrase, ‘the juice is worth the squeeze’!
It’s no secret that technology has the power to create value in the supply chain – in helping companies make smarter decisions faster. But despite the promise of AI and machine-learning and all the automation that may exist, you still need the human capital and the talent that is able to take the complicated world of the physical supply chain, put it within the constructs of a digital model and leverage the power of mathematics in playing out different scenarios.
It requires a deep understanding of your operational processes and your business combined with some fairly advanced mathematical skills. You don’t have to be a PhD to do it, but you do need to understand the constructs of the data model and how to use software to represent those physical complications. That’s a skill set that is not only very different from what was needed in the past but is also currently deficient.
The skills gap has been growing over time and there needs to be real focus and emphasis on understanding ‘what good looks like’ in the supply chain professional of the future and on finding ways to seed and cultivate these new skills.
While every supply chain is vulnerable to unforeseen threats and force majeure, understanding the weak points allows corrective action to be taken and/or robust contingency plans to be put in place. The simple act of using technology to visualise your end to end supply chain will likely reveal previously unidentified fault lines and you can also use predictive analytics to expose some of the hidden risks, both in the near and long term.
As ‘real-time’ as everyone wants to be, in supply chains there are lead times – if you’re bringing in product or raw materials from Asia or China into Europe on container ships, there’s still a 30 day transit time from your origin in Shenzhen to your warehouse in Northampton, so understanding and being proactive about risk and getting ahead of the curve to handle those lead times are very important.
The evolution of the supply chain industry over the last 20 years or so has been primarily focussed on process automation – as a result, few supply chain systems have been architected to enable what-if analysis in a robust way. Demand planning is a good example – while these systems may have created a lot of value in driving forecast accuracy, the assumptions you made when you implemented those systems 20, or even five, years ago are no longer valid in today’s economic environment. However, these systems were not designed to question or interrogate those inputs, parameters and assumptions that were made.
The ability to do what-if analysis is a critical requirement for making good business decisions. What-if analysis allows users to explore multiple scenario options, compare outcomes of different alternatives and ultimately make better supply chain decisions.
In conclusion, while technology can’t show you what the future holds, today’s leaders are investing in the data foundation, integration, talent and robust what-if analysis capabilities to accurately answer bigger questions more often and make bolder decisions with confidence.