Monday, May 28, 2012

Automated Store Replenishment – Volume VIII

Studies have shown a reduction in storewide inventory of 25% delivers 2-3 percentage points better levels of service. Using these figures as a yardstick, one could extrapolate a 70% reduction in inventory might result in 8-9 percentage points increase.

In addition, several years ago, a major beer distributor did a controlled study that proved, reducing inventory levels to a 1.5 delivery cycle increased sales by 40%, mainly because it reduced the clutter in the stores. The conclusion drawn from this results in a triangle, with ‘customer service’ on one side, ‘increased sales’ on another, and ‘reductions in inventory’ on the third. We could state the obvious that under certain conditions, ‘reductions in inventory’ are directly proportional to both ‘sales’ and ‘customer satisfaction’. But what are those certain conditions?

I was able to identify a few dozen without trying very hard. It only makes sense that we don’t need to buy more inventory than we can sell, except in the case where we can make a special deal on something, or receive a insider’s tip that sales on the item are going to explode, or a severe shortage is just around the corner.

Holding an item’s volume to 1.5 delivery cycles means that if you sell five units during a delivery cycle, the correct amount of inventory to have at the beginning of each cycle (taking into account seasons and holidays) is 5 units, plus 50% more (rounded up to the integer) for safety stock. In this case, (5 * 1.5) =  8 units. The last time I heard, convenience store inventory was moving at an overall rate of only 0.22 times per day, and this low figure is indicative of the fact that 70% of your inventory is either dead, or moving so slowly it brings down the overall movement rate to unacceptable levels.

“But I can only order this product in units of thirty-six?” Then the best option is to keep the excess inventory stacked neatly in the storeroom. It will take up much less space because you can stack boxes more efficiently, organized by the sections on the sales floor. When the sales floor level reaches a predetermined level, the computer alerts the person in charge of the section to impress the level on the floor back up to eight, and when the overall quantity reaches eight, ten, twelve (whatever), the system orders another box of thirty-six.

Can it be done? Absolutely! But a system like this will not work in an environment where the sales floors and storerooms are packed with excess stock. In addition, you will need to manage an integrated inter-organizational relationship with suppliers and trading partners with risks and rewards shared by everyone involved in the supply chain.

When your managers attend trade shows and are persuaded by manufacturers to buy more product than they can possible cram into their stores, it creates a log-jam in the supply chain that is impossible to manage and oftentimes abused. The manufacturer may perceive this as an advantage, however all he is really doing is sacrificing a sustained and efficient movement of products for a temporary spike in sales. 

This is the kind of push-based system that is doomed to failure. Once this kind of relationship is established, it is almost impossible to break. Eventually, the retailer goes out of business and the supplier loses the sales channel completely. Suppliers, forecasting future sales based on these kinds of relationships, are playing with fire.

Push-based systems may be convenient, but they are also deadly, while a pull-based system is impossible for the human mind to manage. So how do we get there for where we are today?

No comments:

Post a Comment