Monday, May 28, 2012
Automated Store Replenishment – Volume IX
The process itself is not beyond the capabilities of the human mind. I visit single-store operators all the time who do a fairly good job of it. There are exceptions however. I gave a copy of my first book on inventory control to a retailer up the road three years ago, and I’ll bet you he has asked me for my phone number twenty times since I gave him that book. He is a procrastinator. He knows he needs to do it, but he hasn’t found the time.
We start with the objective of having the right amount of inventory in the right place at the precise moment the customer is ready to buy. We make it profitable by limiting the amount of inventory to being no more than is required to meet customer service levels, and limit Out Of Stocks (OOS) by building a little ‘safety stock’ into the equation.
For example, our ASR keeps a running calculation of inventory movement over and infinite period of time, and from that table we have access to an infinite array of sampling periods—yesterday, during the past week, the past two weeks, past month, past year; and we have the ability to analyze spikes during seasons and holidays. From that, we can come pretty close to getting an accurate estimate of how many items the store will sell before the next delivery cycle occurs.
This is a lot of fun: http://www.crsys.com/Test3.HTML?Store=AAZ01.0115&UPC=040000001027 Change the last twelve digits to any UPC code. (Observe upper/lower case). If this store sells it, you can view its ninety-day history. Move your cursor over the graph to get precise figures. The last sale occurred today at 9:01.47.88 AM, the average turn rate on this particular bar of candy is 4.0537, the current retail price is $1.09, the moving-average cost is $0.5778. The next sale of this item will produce EXACTLY $0.5122 in gross profit; and at this precise moment in time (May 27, 2012 at 10:39 CST), they have 125 bars in the store, enough to last for thirty days. They get two deliveries a week—Tuesdays and Fridays. They sold 34 during the last seven days, 74 during the last 2 weeks, 108 during the last 21 days, 140, during the past month and 252 during the past two months. You tell me how many they need to order. Simple, right?
Most current methods used rely on past purchases to estimate current needs. For example, “We received 24 of these last week, so we will probably need to order 24 next week.” And, they rely on the store manager to tell them when the shelf is overflowing.
I will never forget the time I was auditing the first store to implement our Cloud scanning service back in the spring of 2004. During a three-month period, everything I thought I knew about controlling inventory in a retail environment was turned completely inside out.
While I was counting the Orbit Wintergreen chewing gum, Kim, the store manager, told me to look under the cigarette display where I would find more. There, I discovered a sufficient amount of that product to last the store for 1,100 days, and they received a new box every week. Further, Kim said she had repeatedly requested the vendor stop sending her more of that particular product. She finally gave up and just started cramming them under the cigarette display.
As the years have gone by, I would testify under oath, ‘this is NOT an anomaly’. In almost every store I have visited, I have found vast quantities of inventory crammed into every nook and cranny, to the point that every store has at least twice the inventory it needs, if you just go out and look for it. This led to our belief that a twenty-store operator has enough stock on hand to supply forty stores, half their size.