Predict, Not Just Respond

Why Do We Plan?

When it comes to supply chain plans, we plan for a number of reasons. First and foremost, the cost of not planning is extremely high resulting in much higher operation costs and bad customer service. In other words, the recipe for going out of business! Basically, we plan to be predictable, to have visibility and to avoid surprises. But how do you measure your planning prowess? How do you know the difference between an optimized plan and just a doable plan? How can you tell if a plan is better than the other and how do you know a good planner from a novice one?

Once a supply chain has been designed and established, its efficiency is dependent on how you run it? In other words, its “operating system,” i.e. the system that controls it. The goodness of the operating system (better known as the supply chain planning system) can be measured by operations and inventory cost as well as on-time delivery. Factors that contribute to good results are:

  • Accuracy
    • How good the model is in reflecting the real life and complexity of the supply chain
  • Quality of the results
    • What algorithms are used and how
  • Speed
    • How fast does the system respond to changes and produces optimal plans?

A typical, S&OP system is only about 60-70% accurate. It uses the spreadsheet technology of 60s and 70s by use of fixed lead-times, bucketed capacities and pre-defined bottlenecks to generate plans. It is generally fast, but then the spreadsheets are fast too but lack accuracy. Therefore, the question is how can you expect to execute on-time delivery with true visibility, if the plans are not accurate in the first place? The simple answer is you cannot!

In order to obtain plans that are executable (i.e. accurate), it must have a true image of the real world in terms of supplier capability and capacity representation. It also needs to have algorithms and data representations that are designed for optimization in real-time. Finally, it needs to be able to win the trust of the end users by providing results that can easily be validated and give guidance to the end users as to what the decisions are and why. The system should also be able to react to changes in an almost real-time manner. Surprises are a real part of every supply chain. Both on the demand side as well as the supply side, things can go wrong unexpectedly. An optimal system either has already accounted for potential risks and/or it can “repair” the plan to adjust for the changes. Such a system is Sales and Operation Execution (S&OE) which complements S&OP and or supersedes it all together.



Leave a comment