What is Shock Resilient AI Demand Planning and How does it Work?
We mean multiple AI (Reinforcement Learning with human feedback in our case) decisioning agents that are trained on top of digital twins that simulate normal and shock scenarios, and take into account an extensive set of external signals that represent world and industry context from our supply chain Knowledge Graph. There is still room for forecasting models that are part of this decisioning system, but it doesn’t do the heavy lifting. The AI decisioning Agents do.
These AI Agents then recommend decisions along with their impact on the business KPIs. Human planners still have the upper hand and choose which Agent is most applicable to the current context, normal Agent1, or Shock_A Agent, or Shock_B Agent and chooses the recommended action from the selected AI Agent to deploy as is or override fully informed by its KPI impact. The entire system continuously learns and self-tunes based on that usage and feedback. It is always on and it is where smart planners would rather spend their day.
It’s important to note that the external signals have different velocities, commodity prices are daily for every trading day, however, government data (interest rates, unemployement rate per state, CPIs, PPIs etc) is mostly monthly cadence, for example.
Minimize inventory costs while meeting customer demand with Auto-replenishment decision models
Increase and maintain accuracy of forecasting
Forecast, analyze and improve your OTIF scores
Move faster and save time by always having relevant, real-time intelligence for decision making.
Deep learning continuously improves over time and dynamically adapts to your growing business.
AI automates processes and analytics enabling you to focus on customer satisfaction.