For centuries, we’ve figured out most of reality based on three things: observations, analyses, and ideas. The process of collecting information (observing), identifying patterns (analyzing), and devising theories (ideation) has given us a sense of how the universe, living systems, and even basic human psychology works. But of course, there are some things we know better than others. 

For instance, it’s much easier to predict the behavior of subatomic particles in the aftermath of a proton collision than the weather in Tokyo one month from now. We can say more about how antibodies respond to viruses than we can about how global markets respond to changes in trade policy. Why is that?

Weather patterns and global markets are examples of complex things that are basically too open to study easily. They have many variables that affect how they behave, isolating those variables is hard, and those variables interact and produce effects that are stranger than the mere sum of their parts. These days, global supply chains have become some of the most complex entities we deal with.

Global supply chains are affected by geography, politics, weather, supply & demand, infrastructure, technology trends, and management. Knowledge about what your supply chain is doing, and what it will do if certain factors change, is indispensable to business goals, especially in a marketplace where supply chain competency has become a main differentiator between competitors. But historically this kind of knowledge has been very hard to discern because of the inherent complexity involved. But that’s about to change in a very dramatic way—all because of computational modeling. 

Reality as Computer Simulation

In complex problems like supply chain dynamics, it’s important to note that the process of observation, analysis, and theorizing isn’t broken. It’s just been more difficult to execute here than in less complex, closed domains. But cheap and ubiquitous computing power is finally making headway. 

Complex problems need a lot of data to analyze, across multiple factors. Collecting and storing “big data” is easier than ever thanks to cloud storage. The cloud also provides the architecture to analyze data, via large computing clusters can provide massive computational power at low cost. Machine learning algorithms can then feed off of that analysis to develop better empirical models that can accurately predict reality. From simulating bacterial life, to the turbulent atmosphere of Jupiter, to the extinction of the dinosaurs, computer models are digitizing every aspect of reality, so we can understand it all better. Supply chains are an inevitable territory for the application of computational modeling.

Already, for specific aspects of the supply chain, computational models have revealed insights. For instance, a global team of university researchers recently mapped the effects of what consumers buy to the loss of endangered species via supply chain impact. 

Having the ability to link different factors, events, and decisions to consequences will be immensely powerful for supply chain executives and managers. The non-obvious but critical patterns in the supply chain that computational models will unearth will provide insights on everything from risk mitigation to demand execution to smarter production processes. For instance, what might have a bigger impact on a specific high tech manufacturer’s supply chain, an errant tsunami or a tariff on a critical rare earth metal? Computer simulations of the supply chain will provide new ways to assess priorities. For those concerned strictly about the bottom line, these insights mean hard dollar cost savings via new efficiencies and potentially new revenue streams.

The Journey to ‘Digital Transformation’

The ability to simulate, model, and predict real world supply chain dynamics will be the end result of a process that many savvy businesses are currently undertaking, called ‘digital transformation.’ To reap the benefits of simulations, models, and analytics, the first step is to provide the fodder for all those processes—data. 

Remember, the supply chain has a lot of variables that need to be measured. So, the only way to handle all that disparate information is to digitize it at the source, and then store it one dynamic information network. Thereafter, it can be accessed, analyzed, shared, and have its insights transmitted back down to the various stakeholders at the operational level.

Getting that data in one place will be the biggest challenge for companies seeking to reap the benefits of a complete, end-to-end digital supply chain. Today’s supply chains contain many different organizations and information systems. These distinctions essentially create data silos which make any wholesale digitization efforts difficult. But if companies really want to survive in an increasingly complex world, they need to begin the journey toward a completely digital supply chain. That means taking steps to digitize the supply chain, every part of it, by networking together disparate silos. It might appear daunting, but all complex problems do. And yet, if we can digitize our own lifeblood, then maybe the lifeblood of global trade isn’t too far off.