Sales distribution pattern

Table 4.8 Diversity of Sales Distribution Patterns of the CPI and Related Industries, 1987...

With respect to demand certainty, demand is forecasted with bid character and is not stochastic following for example a normal distribution pattern, since demand is influenced by the price development. With respect to demand volatility, demand prices and quantities are not stable but monthly volatile. The total demand elasticity is smaller or equal to 1 with respect to average prices. That means that average prices for total demand can change, if more or less sales quantity is sold in the market. 

In general, the distribution of revenues from the sale of lottery tickets follows the pattern of 50 percent to the winners of prizes, 17 percent for administrative expenses and 33 percent for education. 

Patterns of distribution and customer identification are difficult to obtain from public information. The information used in this chapter comes primarily from the Cefic European Chemical Industry Council and ACC reports. As mentioned earlier, in terms of regional sales of chemicals (Figure 1-1), Asia (other than Japan) accounts for the largest share 30.4 percent of 1.82 trillion total world chemical sales in 2007. 

Failure to understand the underlying patterns and the reasons for variances — Business people generally do not make enough use of the information they have in collective databases to improve the accuracy of sales forecasts. In a typical retail firm, for example, there is data on trade sales and marketing results, retail store transactions, warehouse and distribution center releases, and promotional results, all of which could span five or more years, only a portion of which is integrated into the forecasting process. 

Consider a buyer at Bloomingdale s who is responsible for purchasing dinnerware with Christmas patterns. The dinnerware sells only during the Christmas season, and the buyer places an order for delivery in early November. Each dinnerware set costs c = 100 and sells for a retail price of p = 250. Any sets unsold by Christmas are heavily discounted in the post-Christmas sales and are sold for a salvage value of i = 80. The buyer has estimated that demand is normally distributed, with a mean of p = 350. Historically, forecast errors have had a standard deviation of a- = 150. The buyer has decided to conduct additional market research to get a better forecast. Evaluate the impact of improved forecast accuracy on profitability and inventories as the buyer reduces o- from 150 to 0 in increments of 30. 

Much of this lost demand was due to inaccurate distribution of inventory across sizes. Often, we observed that a particular size in a given style/color accounted for a large proportion of stockouts, while other sizes of this style/color had to be eventually sold at markdowns below cost. To reduce such misallocation within the size distribution of a product, our clustering method could be applied to historical data on sales by size for a product to form clusters of products that had similar size selling patterns. We would then examine the nature of the products in each cluster as a way to understand how size distribution differs by product. For example, this might result in ten distinct size distributions and a definition of the t5q)es of products that had a particular distribution. A new product could be assigned the size distribution of the product type that it best fits. 

---