The three types of supply chain analytics are descriptive, predictive, and prescriptive. Each plays a crucial role in managing your inventory, and modern supply chain software allows you to leverage all three.
Descriptive Analytics are the foundation of dashboards, providing insights into "what’s happening now." This includes summary numbers such as dollars currently invested in inventory, current customer service levels, fill rates, and average supplier lead times. These statistics are essential for tracking operations, especially when monitoring changes from month to month. Descriptive Analytics rely on accurate corporate databases and statistical processing.
Predictive Analytics are most commonly seen as demand forecasts, often broken down by product, location, and sometimes customer. These statistics offer early warnings, allowing you to adjust production, staffing, and raw material procurement to meet demand. Predictive Analytics also predict the effects of changes in operating policies, such as increasing order quantities. They build on Descriptive Analytics by adding advanced statistical processing to detect trends, seasonality, and regime changes. Predictive Analytics for inventory management use demand forecasts as inputs into models that estimate key performance metrics like service levels, fill rates, and operating costs.
Prescriptive Analytics focus on recommending decisions to maximize inventory system performance. They suggest what actions to take next, using Predictive Analytics as a foundation and adding optimization capabilities. For example, Prescriptive Analytics software can determine the best future values for Min and Max for thousands of inventory items, aiming to minimize operating costs while maintaining a desired service level.
The figure below illustrates how supply chain analytics assist inventory managers. The columns show three predicted Key Performance Indicators (KPIs): service level, inventory investment, and operating costs (holding, ordering, and shortage costs).
| Scenario | Avg Full Cycle SL% | Total Avg Inventory Value | Total Annual Operating Cost |
| Live-07-01-2018 | 87.54% | $3,112,697.27 | $665,193.88 |
| 99% All | 99.46% | $3,960,738.11 | $798,439.37 |
| 75 floor/99 ceiling | 92.16% | $2,841,330.61 | $607,944.23 |
| Optimization | 80.71% | $1,030,076.08 | $256,892.54 |
Figure 1: The three types of analytics used to evaluate planning scenarios
The rows present four alternative inventory policies, expressed as scenarios. The "Live" scenario reports KPI values on July 1, 2018. The "99% All" scenario raises the service level of all items to 99%. The "75 floor/99 ceiling" scenario adjusts service levels to a minimum of 75% and a maximum of 95%. The "Optimization" scenario prescribes item-specific service levels to minimize total operating costs.
The "Live 07-01-2018" scenario is an example of Descriptive Analytics, showing current baseline performance. The software allows users to create "What If" scenarios to explore changes in inventory policy. The next two scenarios are examples of Predictive Analytics, assessing the consequences of recommended inventory control policies. The "Optimization" scenario is an example of Prescriptive Analytics, recommending the best compromise policy.
Comparing the three alternative scenarios to the baseline "Live" scenario:
Modern inventory planning and optimization software should offer three types of supply chain analytics: Descriptive, Predictive, and Prescriptive. Combining these analytics allows inventory managers to track operations (Descriptive), forecast future conditions (Predictive), and optimize inventory policies in anticipation of those conditions (Prescriptive).
