More than 81% of supply chain managers say that data analytics will be key to cost reduction in the future—and there are plenty of businesses that bear that expectation out. Amazon, for instance, is currently using Big Data analytics to track supplier inventories and dynamically pair customer orders to the optimal vendors. The result? A reduction of costs by 10-40%. Not only does this help them meet their increasingly stringent (albeit largely self-imposed) shipping turnaround expectations, it gives them a supply chain that’s more resilient as a result of above average levels of transparency.
Of course, for every business that’s putting data analytics to use in the supply chain, there are plenty more that haven’t really jumped on the bandwagon yet. To wit, Gartner found in 2018 that only about a fifth of supply chains were in an advanced stage of supply chain digitization. They also recently found that only a fifth of supply chain leaders reported having a highly resilient supply chain. This is no coincidence—in fact, data analytics can prove crucial to supply chain resiliency. Considering that these uncertain times have made resilience more important than ever, there’s no time to waste when it comes to putting your supply chain data to use.
What Are the Keys to a Resilient Supply Chain?
Before we tackle Big Data and the ways that AI and advanced analytics algorithms can extract value out of that data, let’s talk about what actually makes a supply chain resilient. We’ve seen a lot of examples in the past year of what not to do: e.g. single sourcing, completely eliminating safety stock, etc. But what proactive steps should a supply chain manager take to increase the resilience of his or her operation? Well, step one is to build out strong, diverse supplier relationships. This means that should you put yourself in a position to dynamically source materials from different suppliers in different geographies in a cost-efficient way based on the needs of any given production program. Crucially, this will require end-to-end visibility, such that you can gain insights into your suppliers’ availabilities and capacities before crunch time hits. This way, if there’s a natural disaster or other disruption in a particular region, you’re not caught off guard by shortfalls.
The next step is to take a smart, proactive approach to aligning your buffer and safety stocks with your risk tolerance. The tendency in recent years has been to go as lean as possible, and operating lean is often a virtue, but resilience depends on a measured approach. You might not need enough buffer to get you through multiple weeks of idle production plant time, but you may need more than you have now. Of course, bolstering your buffer stock comes with its own inherent risk—it requires you to up your capital commitments for stock that may never actually be sold and might continue to take up costly warehouse space for longer than you planned. Since untying capital commitments is one way of increasing your flexibility (and in turn your ability to respond to disruptions), this may start to seem like a zero sum game.
Data analytics has the power to change all that.
How Data Analytics Power Resilience
Let's look at an example of how analytics might play a part in making your supply chain less susceptible to disruption: To truly optimize your inventory management—such that you’re lean but not too lean—you need to successfully balance forecasted product demand and forecasted raw goods demand with the structure of your network and the possibility of the unexpected. This is essentially impossible for even the most experienced supply chain planner to eyeball. Instead, it’s the natural purview of advanced predictive and prescriptive analytics. These types of workflows are often powered be technologies like AI and machine learning, and they’re designed to help you create powerful, accurate forecasts and identify potential process improvements up and down the chain.
If you have a steady stream of data coming in from your supplier network, for instance, you can use advanced analytics to forecast availabilities and price fluctuations, such that you can be sure you’re securing materials at the right time and at the right price. If you expand the collection of large caches of data to your larger logistics and transport network—including freight carrier across a variety of modes—you can do essentially the same thing, securing necessary capacity before orders have even arrived. Naturally, the way you’re able to do so is by forecasting product demand and transportation availability more effectively using these same techniques.
As we move from predictive to prescriptive analytics, we can begin to see solutions to the potentially thorny problems of buffer stock raised above. These processes can help you analyze your entire production or logistics network to find areas of waste and redundancy, and they can even help you find the optimal levels of something like buffer stock. Essentially, this enables you to reduce costs across the entire supply chain through efficiency increases, such that you can separate the valuable buffer from anything that’s actually wasteful. Thus, your supply chain’s digital transformation gives you the maneuverability you need to proactively manage risk and increase your resilience.
How to Maximize the Power of Your Data
What we’ve been describing above might sound too goo to be true, but as we get deeper into the era of Industry 4.0 you’ll likely find that these concepts are only scratching the surface of smart supply chain management. Going forward, the most digitally savvy organizations will use user-centered AI to lead planners to the right decisions in real-time, whether that’s choosing the right order volumes from suppliers, making adjustments to a transport network, or setting production sequences in motion. The only question is how, exactly, you can set yourself up to get this kind of value out of your data.
First things first, you need to collect it in a clean, visible way. This means integrating different technologies—from IoT devices and RFID chips to planning and scheduling solutions—in such a way as to make data from multiple supply chain touchpoints available quickly and easily from anywhere on the process chain. Your ability to make this happen will depend in large part on the architecture of your IT ecosystem. A modular, postmodern ERP style of architecture will give you the interoperability you need to avoid silos. From there, it’s just a matter of integrating AI-powered solutions into that larger architecture. As long as you have the flexibility within your IT to make this a reality, you can leverage that flexibility into smarter planning and scheduling—as well as smarter disruption and change management.
In this sense, the true backbone of a resilient supply chain comes down to your technology. Gone are the days—if they ever existed—when planners could successful manage costs and maintain a lean supply chain architecture using a spreadsheet and calculator. Today, to create a highly visible, highly flexible, and highly resilient supply chain, you need to start with smart technology integration.
