Operations Entropy, the bane of planners

All supply chains cope with numerous seemingly unpredictable events – everything from social trends, to customer order spikes, to late materials. While these external drivers are important, we’ve found that internal drivers of disruption are even more numerous and often not fully recognized – creating an execution blind spot.

So, what are internal drivers of planning disruptions? Many of these lurk within the detailed weekly plans over the 12-weeks that drive supply chain execution. Traditional demand and supply planning tools are not focused on this horizon. Examples of these issues include:

  • Extremely high demand error at the SKU, distribution center (DC), and week level
  • Instability in the master production schedule
  • Lack of adherence to factory schedule frozen horizons
  • Extremely high error in simple lost sales projections that drive deployment and production changes

These internal issues are leading to “blind spots”; up to 70% of total exceptions can be missed. In other words, the planners have a limited scope on most of the exceptions. We call this planning challenge Operations Entropy.

Before you reach for a chemistry textbook, we define Operations Entropy as a kind of fog that clouds decisions happening every day in complex supply chains and distribution networks. It leads people to believe that there are random, unpredictable, events lurking around every corner.

The result is unplanned deviations, defects, excess inventory, empty delivery trucks, and unfilled orders; all tangled up in a sea of alerts and noise. No time to root cause because the next “seemingly” random event is around the corner.

2 metrics to help you see clearly through the fog

As a supply chain professional, your priority should be about metrics. This is where the challenge starts. It’s likely that you are not measuring the underlying drivers of the execution horizon challenge – the weekly entropy.

  1. Weekly demand accuracy at SKU, DC, week level:
    Many have questioned the validity of measuring at this detail, but this is the “signal” driving your inventory deployment and production exceptions. It’s likely that your error at this level will be somewhat shocking. The weeks matter and it’s possible to make huge gains. Don’t be surprised if your weekly error is a double monthly error (for instance, 40% MAPE monthly and 80% weekly).
  2. Potential lost sales accuracy:
    Do you measure the accuracy of the potential lost sales reports that your teams use to manage exceptions? Most companies do not. You can get an initial idea of this accuracy by taking a report from 4-6 weeks ago and seeing how it performed. Take the predicted lost sales for this week and compute the error from 4 weeks ago and a few more weekly lags. Again, the error is likely to be much higher than expected. Like the previous metric, this error could be up to 70%.

Armed with this information, start to understand where the error is highest and attempt to find root causes. These basic metrics will tell you a lot.

Done that, what now?

Noodle.ai is on a mission to create a world without waste. Our AI-powered product suite, designed specifically for supply and manufacturing, empowers companies to eliminate the most significant waste culprits: excess inventory, product defects, unplanned downtime, and unfilled orders.

Try our quiz: FlowOps Maturity Curve for Supply Chains.

Learn more: Explore our products Demand Flow, Inventory Flow, Production Flow.

Request a demo: Experience a live customized demo of our product suite.