Coping with COVID-19 in supply chain planning
Over the last two weeks, we have all shifted rapidly to working from home, requiring quick adaptions for working and family lives. Most supply chain planning teams are now coping with a hard reality – legacy planning and ERP systems were not designed to cope with this uncertain environment. No one has a crystal ball that can predict how this tumult will unfold. We do have new data science and AI/ML capabilities which can improve visibility and begin to help stabilize supply chain plans and make teams more effective.
First, traditional statistical models are based on sales history and not able to effectively react to a huge change in sales pattern. Instability and obvious issues in the outputs are leading many companies to turn these models off and revert to manual practices. This is not efficient in a situation where your business outlook is very uncertain and constant changes will be the reality. Even where sales are very high, many CPGs are struggling to get a handle on consumption versus stocking/hoarding. There are parts of both and the dynamics are likely to change as the crisis unfolds.
Second, many distribution planning and manufacturing planning systems require very granular forecasts to operate – typically SKU/DC level– which can mean millions of forecasts for many companies. The old response is to double down on spreadsheet-based approaches. However, with everyone working from home, recreating a planning war-room with dozens of people has been difficult and constant changes in outlook cannot be quickly incorporated.
An AI-Based Approach
Adding a product-based data science and AI/ML capability can be the glue to hold things together. Noodle.ai data scientists have been using the Athena Supply Chain AI platform to develop “good enough” demand baseline plans. As we get a few more weeks into this, we are seeing evidence that AI/ML predictive models will start to help improve demand plans and provide the level of detail that the legacy supply planning systems require. This forms the basis of a virtual AI mission control center and will help planning teams to be more productive by focusing on the right exceptions. These are difficult times, but we are optimistic that these new capabilities in an AI/ML based supply chain suite of applications can play a significant role in navigating the next few months and year.
- Atlas, Noodle.ai’s Machine Learning (ML) Framework Part 2: Design Premise & Architecture
- Experiencing Supply Chain Blindspots?… Enter Athena Insights
- Atlas, Noodle.ai’s Machine Learning (ML) Framework Part 1: Challenges With Building AI Applications
- Selling Certainty in an Uncertain World: Why Noodle.ai is a “Top 15 Startup to Emerge Stronger from the Crisis”
- The Marketing of Industry 4.0: The Hype Cycle Ends Where Reality Begins