Results-driven Decision-making for Supply Chain Since 1971

Q: Why do I care about Machine Learning for my supply chain?

A: Because of the staggering number of decisions, variables, and trade-offs you have to make each day.

With literally hundreds of thousands of SKU/Location combinations including your own products at company, distributor/retailer, customer and vendor facilities, it is impossible to know which orders or changes in ship date, due date, quantity, etc. are meaningful. To make matters worse, without ML, it is likely your employees don’t know about a problem until it has already happened. Here are examples of ML’s ability to anticipate and measure problems before they arise with critical component shortages:

Prospective Risk Management Methodologies

  • Precedented
    • Utilizes directly-observable events to use embedded Supply Variability optimization in the GAINS model
    • Can apply Supplier Variability data to target operations as proxy to quantify max buffer (which will often be smaller than rule-based estimate)
  • Unprecedented at the Manufacturer Level
    • Respects risk without over-mitigating (especially after, rather than prior to, events)
    • Quantifies magnitude of risk & mitigating buffers utilizing parameters such as:
    • Time-to-Survive (TTS)
    • Time-to-Recover by ‘Node’/Supplier (TTR)
  •