For years, the application of predictive analytics in supply chain management (SCM) has been described as “transformative,” a “big opportunity,” the “new business intelligence” and even “the holy grail.” While it is agreed that predictive analytics can be all those things (and more), there is confusion on where and how predictive analytics should be applied.
The basic purpose of supply chain management is to control the manufacture, storage, transportation and sale of goods and services to most efficiently meet customer demand. You need to have it when they want it and keep nothing else. Predictive analytics involves processing and consuming a large amount of data to gain insight into possible future scenarios and their potential outcomes.
Since this is a primer, I will focus on only two major areas where predictive analytics can be applied to SCM: customer demand forecasting and inventory optimization.
Customer demand forecasting
Demand forecasting is defined as the prediction of future consumer demand for a product or service. In the Deloitte survey Supply Chain Talent of the Future, demand forecasting was listed as one of the top “fast-evolving technical capabilities” in use or expected to be used in the supply chain.
Demand forecasting isn’t new. However traditional methods, as noted in a BCG article, are often powered by limited data, are time-intensive and use outdated forecasting models that are typically of the one-size-fits-all variety.
The improvements to demand forecasting using predictive analytics are far-reaching:
- Insight: The machine learning employed in predictive analytics models allows for large amounts of structured ERP and SCM data to be processed with seemingly disparate data, such as consumer sentiment data (Facebook, Twitter, Pinterest, Instagram), macroeconomic indicators (GDP, unemployment, Leading Indicators Index), IoT device data, demographics, weather and other domain-specific factors (production lines, engineering changes). It’s not just about collecting data, predictive analytics enables actionable insight into how, when and why customers make purchases.
- Speed: Near real-time collection of data combined with machine learning allows predictive analytics to be calculated within minutes or a few hours, rather than the 24/48-hour batch processing cycle that is often used today.
Pier 1 Imports has a great case study of how they’ve used predictive analytics to understand customer shopping habits across 1000 neighborhood stores, and how they acted on those insights. While supply chain management isn’t specifically mentioned, the case study provides interesting insight into how predictive analytics impact their decision-making processes.
By understanding and forecasting customer demand, another major area of applying predictive analytics in supply chain management becomes relevant: inventory optimization.
An article in SupplyChainDigest gives a solid definition of inventory optimization:
…having the right amount of inventory, in just the right places, to meet customer service and revenue goals — but no more than that.
Inventory optimization helps reduce inventory distortion, a challenge that stems from out-of-stock and overstock inventory situations. A study by IHL Group, a global research and advisory firm specializing in technologies for the retail and hospitality industries, found that inventory distortion costs retailers nearly $1.1 trillion globally.
Traditional methods of inventory optimization include adjusting inventory retroactively — reacting to customer purchase habits that have already occurred. This is further compounded by segmented IT systems set up to meet isolated needs, resulting in decentralized and short-sighted inventory decision-making processes. This method often results in significant issues throughout the supply chain.
At one point, Southern States Cooperative, a U.S.-based agricultural supply cooperative, had more than 1000 retail distribution points and recognized slow-moving inventory (SMI) within its supply chain:
To address this issue, Southern States implemented a multi-faceted project that leveraged data mining, predictive analytics and a transition to a centralized inventory management system. Even in the early stages, they noticed immediate improvements:
Aside from the reduction of inventory while maintaining sales volume, Southern States realized other benefits of employing predictive analytics in its inventory optimization strategy:
- Actionable data delivered to multiple operating units to improve inventory management
- Greater insight into seasonality, impacting the timing of delivery of seasonal products
- More accurate sales forecasting, tied to inventory levels, to ensure in-stock positions
- Cultural shift in the use of information to improve the business — fewer decisions based upon gut feelings, more based on the analysis of the business.
Predictive analytics will help you identify trends, understand your customers’ purchase habits, predict purchase behavior and drive strategic decision-making. If you’re not employing predictive analytics in your SCM strategy, put away your complicated spreadsheets and isolated databases and start considering a transition today.
At Blueprint Technologies, we’re experts at connecting advanced data science and analytics with supply chain management to help organizations best serve their customers. Let’s start a conversation today about your specific needs.