An ever-expanding set of global challenges has meant that in every year over the past several years, at least one company in 20 has suffered a supply-chain disruption costing at least $100 million, according to a recent report by McKinsey & Company.
In commodity markets, the buying and selling of raw or direct materials have always carried risk, even before the global pandemic. Following the outbreak, commodities supply chains were found to be particularly vulnerable, hampered by inconsistent manual processes, disconnected systems and inaccessible spreadsheets. Lacking visibility into the causes of supply-chain disruptions, commodities producers, traders and consumers were unable to respond to a rapidly changing landscape.
Companies that produce, procure, or trade raw materials are eager to realize the benefits of well-defined processes supported by digital technologies. For these businesses, the inability to identify risks and respond quickly to disruptions can be especially costly.
According to a McKinsey survey of 60 supply-chain executives in the second quarter of 2020, 73% encountered problems with suppliers, 75% struggled with production and distribution, and nearly half experienced delays in planning and decision making. More telling, at a time when workforces and supply partners were sequestered remotely, a stunning 85% of companies reported struggling with insufficient digital technologies to support key processes.
For the past several years, many companies have turned to supply-chain solutions designed for commodities to support buying, selling, moving, storing and processing, while managing the risks associated with those functions. While such applications are critical to evaluating and tracking assets, the need for specialized supply-chain support — particularly at critical transfer points across stockyards, bulk terminals, warehouses and port terminals — remains.
New technologies such as the internet of things, artificial intelligence, machine learning and 3-D modeling are making it easier to track and manage raw materials. They allow companies to begin answering critical questions about the status of their goods, from farms and factories to storage facilities, ports and processing plants. But without an integrated view across the commodities lifecycle, those technologies will fail to deliver the level of visibility that global enterprises need to build resiliency, manage assets more intelligently, and make decisions in real time as conditions on the ground continuously change.
Building resiliency across the raw-materials supply chain is a critical priority today. The ability to respond more nimbly to unanticipated disruptions, and recover faster, has never been more urgent.
Producers, buyers and traders are adept at managing a diverse portfolio of strategies to mitigate risk-optimizing production and pricing, by participating in exchanges and designing contracts and other financial instruments to hedge against loss. But as the lessons of the pandemic have made clear, risk is embedded at every point across the commodities lifecycle. A mill with capacity but no visibility into whether its suppliers will be able to fulfill orders is unable to secure alternative sourcing. A lack of storage capacity to receive shipped goods prompts an expensive scramble for warehouse space. Standardized processes for stockpiling make it difficult to verify quality precisely enough to blend materials with confidence.
Organizations seeking to master this volatility will need a more granular view across their supply chains. The lifecycle for raw materials involves a global network of producers, transportation providers, storage and processing facilities, traders and consumers. At each point in the chain, people, processes and technologies need to be seamlessly connected for up-to-minute decision making about critical assets.
Businesses operating in direct materials need to push beyond standard stockyard management and warehouse-automation practices, and embrace precision modeling tools that deliver deeper visibility into what’s happening on the ground. Armed with a more granular view of equipment and goods, they can dynamically adjust operations in real time to increase throughput, manage quality more flexibly, and simulate task planning for faster decision making.
Storage, blending and transfer operations are among the most critical touchpoints in the upstream and raw-materials supply chain. These activities are supported by efficient and specialized bulk terminal automation solutions that, ideally, provide visibility into assets, improve facility utilization and, ultimately, increase gate throughput.
Three-dimensional modeling brings new levels of visibility and control to stockyard and terminal operations. Site operators can see all their assets in a 3-D “fly-around” user interface, to plan and execute receiving, transfer and loading tasks more effectively. With 3-D modeling, machines can work in greater proximity — within a meter of each other rather than 10. In addition to supporting more efficient performance and increasing throughput, 3-D modeling effectively enlarges the footprint of the stockyard, allowing companies to carry additional inventory without expanding facilities.
The science of stockpile management continues to evolve. But even a well-stacked stockpile presents challenges for commodities producers, who must constantly weigh quality against throughput in order to meet contract requirements.
Without real-time visibility into the stockpile, producers are at risk of underdelivering on contracts and triggering penalties. To avoid those penalties, many producers err on the side of overdelivering, often exceeding quality parameters by as much as 2%.
Today, 3-D modeling allows organizations to manage their stockpiles more precisely, by scanning materials as they flow into the stockpile, then modeling that data and delivering a digital twin of the stockpile to operators. The models can calculate quality issues in real time as goods are stacked, reclaimed, blended and delivered to buyers. In other words, as stockpiles are being stacked from multiple sources, as those layers are being reclaimed by various means and blended for delivery, stockpile operators have a precise — and even predictive — understanding of stockpile quality from receiving to delivery.
Such insights have multiple applications. Rather than relying on a pre-programmed sequence of events to automate processes, operators can apply unique control techniques, running several jobs on a conveyor belt and dynamically adjusting them on the fly. No longer must they run the belt for a fixed period of time, clear it, and begin again.
Predictive simulation uses machine learning to ask and answer new questions about stockpile management: How many stockpiles are optimal for any given stockyard or bulk terminal? Which elements are best combined to deliver more value to a customer? What kinds of materials should be kept separate? Is it more optimal to blend as you stack, or to blend onboard a vessel? What would it mean to follow the lead of consumer brands, and deliver more “personalized” or bespoke silos?
Answering those questions will require access to significant amounts of data from across disparate sources. Increasingly, companies that want to set themselves apart will need to rely on a connected network of suppliers, sellers, buyers and traders, not simply to optimize operations, but to reimagine how commodities markets will continue to thrive.
Disruptions driven by the pandemic have exposed the need for commodity managers to master business processes across their supply chains. At the same time, they must develop a deeper understanding of the assets — people, processes and technologies — that move commodities across that chain. The pandemic helped us to understand the critical importance of just-in-time management, especially at the most vulnerable points in the supply chain.
Manav Garg is founder and CEO of Eka Software Solutions.
Timely, incisive articles delivered directly to your inbox.