Distribution and Supply Chain
The biggest issue in distribution is to achieve having „the right product in the right quantity at the right place“.
In general, customers have limited tolerance to product unavailability – if they are willing to wait for the product to be ordered and delivered, everything is OK. But for products that need to be available immediately, the situation is worse. The only satisfactory solution is to establish some product inventory – a warehouse.
Level of inventory is then determined using a simple formula.
Max inventory level = estimated average consumption during replenishment periods multiplied by an unreliability index.
For example, consumption is 5 units per day, supplier replenishment time is 5 days and sometimes shipment arrives a day later or supplier ships 20 % less than we ordered (i.e. 20 % longer or 20 % less). Max inventory level is therefore 5 x 5 days x 1,2 = 30 units.
Organizations usually solve such tasks using warehousing systems and MIN-MAX approach. And we usually perform these calculations for a large amount of items (hundreds or thousands) simultaneously. For every item we need to define the MAX value (max inventory level) MIN value (replenishment order value), safety stock and supplier replenishment time. These values are rarely changed.
When we define parameters and very carefully manage our inventory, inventory levels usually stabilize on certain value, but no matter how hard we try, we end up in situation where overall inventory levels are reasonable, but some items (approx. 10 % – 25 %) are missing (“stock-outs”) and a fairly big portion of items (approx. 30 %) is laying idle and not sold (“surplus”).
Both stock-outs and surpluses harm the company – surplus ties up money and stock-outs mean lost sales and profit (or margin).
What is the root cause?
On one side, there are customers – consumption cannot be forecasted precisely, customers change their preferences, advertising and discounts on certain products affect sales, products are sold out and cannot be re-stocked in time. That is how stock-outs emerge and cause lost sales.
On the other side the same change in customer preference leads to surpluses in other products. More importantly these surpluses are often exacerbated by suppliers and their approaches to stock replenishment. Usually they try to “flood” the sales channel with their product under the notion of efficiency (campaign = better price, full truck = lower transportation costs, minimal production/shipment batch = better unit costs etc.) Therefore there is often a much larger amount of stock in the pipeline than can be realistically sold during replenishment cycle. That is how surpluses emerge.
Moreover, both sides (customers and suppliers) are quite independent. Increased consumption may not occur in times when the purchasing conditions are the best and vice versa. So the gap between consumption and replenishment widens thus creating more stock-outs and surpluses simultaneously. In general, distribution systems suffer from very high degree of variability.
Because the MIN-MAX management or forecasting approach utilizes static data, the situation gets even worse. We are able to replenish inventory to defined MAX levels, but we are unable to react quickly to sudden changes in demand. It is true that many companies are working with MAX levels and adjust them according to demand, but due to the huge number of items, the change occurs usually once in a few months (typically quarterly or semiannually). But changes in consumption and replenishment are much more frequent (days to weeks) and therefore products might be replenished correctly, but according to incorrect MAX inventory level.
What is the solution?
To resolve this puzzle, we need to understand that positive change cannot be achieved through the same means that caused the problem. We cannot achieve better results by improving MIN-MAX management or introducing better forecasting method (random demand cannot be precisely forecasted). But we can change the way we restock our inventory – from “push” to “pull” replenishing – by not deducing future consumption from forecast (forecasts are valid for global demand and long-term decisions, but will not tell us how many units customers will buy in the short term so we can order the right amount in a 2 day replenishment cycle), but from actual consumption.
This approach is based on the same formula for inventory level as “inventory push” approaches -
max inventory level = estimated average consumption during replenishment period multiplied by unreliability index – but uses it in a different way.
The “Inventory pull” approach does not use estimates (forecasts) for definition of upcoming consumption. Rather it uses data about recent consumption (it is very precise), typically sales data from past few days, and based on that (using a non-deterministic algorithm) defines:
• Quantity of a product to order or replenish in a warehouse
• How urgent the replenishment is
• What products MAX level should be increased or decreased
This approach allows changing the replenishment value more frequently, but what is crucial: the target inventory level is always closest to actual consumption in the period. This fact alone significantly reduces opportunities for stock-outs and surpluses to emerge. Reducing stock-outs leads to increased availability of our products to customers, reducing the surpluses leads to decrease in inventory value and free cash previously tied in slow-moving inventory.
These features also support introduction of the basic rule: “Order daily, replenish often and regularly.” It is possible to work with suppliers and achieve even faster inventory replenishment in later phases of the implementation of the pull approach. This leads to further decrease in inventory levels without compromising availability.
By implementing TOC Pull Replenishment methodology (“inventory pull” replenishment approach based on Theory of Constraints) organization can achieve one or more of these benefits:
• Increasing availability of products to customers up to 100% without increasing inventory levels
• 20% increase in sales without any additional advertising or marketing costs
• 30% - 50% reduction of inventory levels without a decrease in sales
• Significant increase of inventory turnover, usually 4 – 8 times
• Significant increase in ROI of working capital, typically multiples
These benefits occur even if the organization has already implemented and uses MIN-MAX efficiently (in all variants) to manage its inventory.
TOC Pull Replenishment can be set-up in matter of weeks (3 – 8 weeks), depending heavily on the capacity of internal IT or vendor of the system the organization is using for distribution management. Full implementation of TOC Pull Replenishment lasts about 6 months. Benefits start to emerge already during implementation phase. Benefits fully materialize after one to three of (the longest) inventory replenishment periods.