The Wilson Formula
The Wilson formula is a traditional method for determining the order or
production quantity if you know the total consumption during a period of
time. The formula assumes that the only costs entailed are a warehousing
cost per stock keeping unit and a one-time cost every time an order is
placed, known as administrative re-ordering costs. The formula tries to
find an optimal balance between the two costs to minimize the total cost,
which is known as the economic order quantity (EOQ).
In order for the Wilson formula to work, a number of conditions have to be met:
The values are hard to determine, and even harder to
The warehousing cost in the Wilson formula is connected to the value of
the article. This is misleading, because it costs more to handle e.g. a
bunch of pipes than one fitting. The administrative re-ordering cost is
hard to determine, and is also dependent on the type of article. For
example: is the transport cost the same if a construction company orders
100 units of insulation wool as when it orders the same number of
screwdrivers?
Because the values are usually dependent on the type of article, a large amount of data needs to be defined. The big problem is maintaining these huge amounts of data, which usually means they are not updated. In other words, you navigate using an obsolete map. The question is, was it ever correct?
Using the formula is an example of suboptimization. The desire to reach an optimal solution to a local problem steals resources from the whole.
We suspect the reason is this: When the total cost is optimized, the graph is very level, which means among other things that a 50% deviation from the optimum only affects 8% of the total cost. Since the formula produces nearly the same result regardless of how the figures vary, there is a feeling of safety, which is reinforced by the scientific aura of the formula.
This way, you can keep following the well-trodden path - and where it will take you is something few consider.
The conditions are not met
In order for the Wilson formula to work, a number of conditions have to be met:
- Demand is constant and continuous
- The lead time for receiving ordered goods is constant
- Administrative re-ordering costs and warehousing costs are constant
- The order quantity does not need to be expressed as an integer
- The entire order quantity is delivered to the warehouse on the same occasion
- No shortages allowed
- The price/cost is independent of time requirements and ordered quantity
The values are hard to determine, and even harder to
keep up-to-date
The warehousing cost in the Wilson formula is connected to the value of
the article. This is misleading, because it costs more to handle e.g. a
bunch of pipes than one fitting. The administrative re-ordering cost is
hard to determine, and is also dependent on the type of article. For
example: is the transport cost the same if a construction company orders
100 units of insulation wool as when it orders the same number of
screwdrivers?Because the values are usually dependent on the type of article, a large amount of data needs to be defined. The big problem is maintaining these huge amounts of data, which usually means they are not updated. In other words, you navigate using an obsolete map. The question is, was it ever correct?
Suboptimization
Many attempts have been made to minimize the effect of the unreasonable conditions of the original formula. For instance, there are amendments that handle shortages, differences in lead time and differences in demand. The problem is that these new formulas require even more data that is hard to collect, and the results are often the same as with the original formula, while the amount of work spent on administration and gathering information is significantly increased.Using the formula is an example of suboptimization. The desire to reach an optimal solution to a local problem steals resources from the whole.
Why is the formula used in spite of the unreasonable conditions?
We suspect the reason is this: When the total cost is optimized, the graph is very level, which means among other things that a 50% deviation from the optimum only affects 8% of the total cost. Since the formula produces nearly the same result regardless of how the figures vary, there is a feeling of safety, which is reinforced by the scientific aura of the formula.
This way, you can keep following the well-trodden path - and where it will take you is something few consider.