Recently, supply chain professionals have recognized that better data collection and increased computing power can track sourcing, scheduling and routing better and faster than any human. Applying big data to thorny supply chain problems is still an emerging art as companies adapt their internal processes to rely on algorithms rather than rules of thumb. Here’s what you need to know to understand how big data is changing the supply chain and improving efficiency. Big Data Definition There are as many definitions of what constitutes big data as there are experts with opinions on the topic. What they all agree on is that big data is more than simply a large data set. The most common definition of big data includes three characteristics:
- Volume: The starting point for big data is a very large data set. While there is no agreed-on threshold, big data volumes are measured in terabytes, petabytes or exabytes.
- Variety: Most business intelligence and data analytics tools work only with structured data — the type of orderly data found in ERP, PLM, CRM and SCM systems. Yet most of the world’s data is unstructured — i.e., data found in documents, presentations, social media, point of sale (POS), weather patterns and news reports. To qualify as big data, the data set must include both structured and unstructured data, and a big data “engine” — a tool or application that can make sense of both types of data.
- Velocity: Big data moves quickly. The engine must be capable of accessing, organizing and finding trends in data that change or update frequently, often in real time or near real time.
Pattern Recognition, Trends and Algorithms Big data engines excel at drawing inferences and recognizing patterns and trends that are so subtle — or so far reaching — that it is difficult for humans to recognize them quickly or at all. A big data engine might notice a certain hashtag or even a word beginning to trend on social media, and use that to generate a prediction of increased sales for a product or category. This early warning would provide the manufacturer with time to mobilize the supply chain to increase supplies ahead of the surge in demand. All trends reach a peak and then begin to lose momentum, and big data can provide insight ahead of the curve that enables the manufacturer to reduce production volumes and inventories throughout the supply chain. The insight provided by big data can improve business outcomes in a variety of ways:
- Big data enables the manufacturer to increase volumes quickly to ensure sufficient supplies. By placing orders for components and raw materials early in the cycle, the manufacturer avoids paying premiums for short lead times and rapid delivery.
- Because the manufacturer has visibility into demand, it can ensure there is ample inventory to meet rising demand, thereby maximizing revenue and keeping customers happy.
- Visibility into inventory at every stage of the supply chain, including in transit shipments, allows the manufacturer to route materials to the point where they are needed most. In turn, it helps meet demand while reducing overall freight and transportation costs.
- By recognizing when the trend is ready to crest, the manufacturer could cut back on inventory to reduce waste and obsolescence. Early insight enables the entire supply chain to react quickly, cutting inventory at every stage and reducing overall supply chain costs.
Visualization One of the most important aspects of big data analytics is the ability to provide easily digested visualizations of the trends and predictions. This makes it simple for humans to grasp the details, see the trends and act rapidly and confidently. Users don’t always know what questions they should ask, nor do they have time to wait for IT to generate new reports. They need insight immediately, presented in an easy-to-understand manner, making visualization one of the most important aspects of big data usage in the supply chain. Supply Chain Benefits Traditional supply chain applications were very inward focused. They concentrated on optimizing the manufacturer’s results, with little regard for the impact on other members of the chain. Today, supply chains are more like carefully orchestrated networks of interdependent cooperating entities. Modern supply chain professionals are holistically focused. They look at the supply chain as a whole, and they seek to optimize results for the entire network as well as the customer. This outward focus requires more cooperation and collaboration across the entire network and insights drawn from a variety of systems and sources. Ultimately, for the modern supply chain, big data should be a requirement — today.
