Final Wac Ll Bean
Content
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
WAC L.L. Bean, inc., a catalog business is unable to mach demand with supply and bearing additional cost of $11 million because of lack of stock and $10 million because of having wrong inventory. These problems are occurred because they are not forecasting exactly what are their customers demanding and how much quantity is demanded. L.L. bean usually place order to their vendor twelve or more weeks before when their customer receive their catalog, it is because the lead time for the production is eight to twelve weeks and it is unable to change the commitment after that. Pattern of early season demand is forecasted by deliveries against a commitment in order to make prediction about end season demand. So it is a challenging situation to forecast demand in this situation. The forecasting problems are very important for L.L. Bean and it is very important for them to find an immediate solution of these problems because there marketing philosophy is " sell good merchandise at a reasonable profit, treat your customers like human being they'll always come back for more." so their customers are very important for them and if they are out of stock, they might lose their customers and if they sell their excess stock at discount price while liquidating, then it would be totally again their philosophy of "sell good merchandise at a reasonable profit." L.L. Bean's marketing approach is direct marketing because they access their current and potential customers by sending them their catalog and then their customers order. Thus, they create demand. This approach had been successful because they reached to their customers directly.
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
Although the demand of a customer is very hard to predict but there are also some prominent mistakes observed related to demand forecasting of L.L Bean which might be cause of the problems. Their product line is like a hierarchy which is divided into merchandise groups, demand centers, item sequences and items and they forecast their demand at item level and in demand forecasting there are two major steps, First they forecast the "frozen forecast" by book in this process, 1st they arrange the items and rank them according to expect dollar sales, after that they assign dollar to each item according to its rank then they use role of thumb to check that if it feels good or if it makes sense. after that they make adjustment if any new item is added to analyze if the item is creating incremental demand or stealing demand of other items. Thus, forecasting is done book by book and item by item and finally frozen and handover the inventory managers. if we look at the whole process of making "frozen forecast" we will come to know that they commonly use the rule of thumb. This was acceptable in the early ages but it is unacceptable in the era of improving technology where the mean of advertisement is changed because of internet and global competition has been started and customer's demand has become highly unpredictable especially, if Scott Sklar, a buyer of men's shirt for L.L. Bean, want to add a new item he would have to observe the trend and gather the information of buyer's behavior related to that "new" item rather than self prediction. with the result of that forecasting, they always observed that the sum of item forecasted remains higher than the dollar targeted to that book so they were uncertain about which item forecasting should be reduced or reduce the whole in some percentage so this was one of the result of under stocking the inventory.
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
In the second step they determine the range of quantity demanded for any (new) item of a product. For this, they take historical errors of each product. value of historical error is obtained by dividing actual demand with forecasted demand for each product. then they compile
frequency distribution of the historical errors then they use that frequency distribution as probability distribution to forecast the range of demand of such item. (see Exhibit 1) They always use A/F ratio for inventory decisions because they want to determine the range of forecast demand for current year and number of units added to stock they do this by multiplying frozen forecast with the A/F ratio. A/F ratio also guides to increase or decrease the frozen forecast for each item by comparing with past data. for instant, if previous forecasted demand exceeds previous actual demand then A/F ratio will decrease current year forecasted demand for that item and vice versa. e.g. previous year forecasted demand is 900 units actual demand is 800 units then A/F is 800/900= 0.88 and if frozen forecast is 1000 then 1000X0.88= 888units. If we look at the process of forecasting range of actual demand for a new, or any other item, item's actual demand would be same like previous pattern of demand but it is not compulsory as Rol Fessenden expressed concern that the there as huge dispersion of errors in "never out" items the dispersion vary at the level of item level like distribution of women's sweeter have more dispersion that than men footwear. it should not be consider that every "never out" item will have same actual demand like all other never out items. The next step is to determine the number of units added to the stock. for this, they consider item's selling price, cost of item, and salvage value or discount price of item and balance the individual item's contribution margin. They calculate the fractile using these costs in order to determine the
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
number of units added to inventory for this they match the fractile ratio with demand probability distribution according to the rank of item it is same as the newsvendor model (see exhibit 2) Mark Fasold, vice president is concerned that number of items which are purchased are usually exceed the number of items because of large dispersion of errors and the errors are used to forecast actual demand so this dispersion disturbs the average value and demand is more forecasted moreover inventory manager makes the policy to gain high profit rather than avoiding losses. In other words he thinks that it is acceptable to bear fewer losses for gaining high profit. The inventory manager should try to balance the actual and forecasted demand rather than adopting that approach. Solution and Recommendations: To solve the company's major problem of matching demand with supply, inventory manager should make forecasting individually of any new item rather than considering "new" or "never out." Because customer's buying behave vary from item to item rather than time of holding that product for the business. For this, they should make the survey to analyze the buying behavior of customers related to that new item. They should do the regression analysis while calculating the incremental demand rather than self prediction and while determining frozen forecast, they should try to consult more and more past data. If the past data is not enough then they should try to gather the data related to behaviors of customer about the items. The company only consults the past data; they can also conduct survey to analyze baying behavior while forecasting frozen forecasts.
Effects of internet: As the competitors of L.L Bean are using internet for the prosperity of their business and they have the salient point of difference among L.L Bean. They should also upgrade their business to the internet. First
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A of all they should contact the service provider to obtain the domain name and register their website. The website should be handled with the networking expert. The website should match the business nature and type. The well designed website would attract more customers. The expenses would be increased but the possibility of revenue would be increased too. It would enable L.L Bean to match the competitors in the open market in the same business. There should be very positive results for upgrading business to internet. First of all the online orders should be made by the customers. The variety could be easily seen in detail and the chances of orders should be increased. The likings and the disliking of the customers could be seen at the spot because of their extended zoom of each product. The cost of opening the new branch or outlet would also be saved. This would be the plus point of the company to reduce the cost and finding the customers at the same place within no time and less cost.
Recommendations related internet era: The experts should add a feature which can calculate the total numbers of clicks hit to every item by the customers. This would enable the management to know the top preferred items and the least preferred items liked or disliked by the customers. The management would be able to focus on those items more which are more likely liked by the customers. The problem which was already faced by the company of wrong predicted items with respect to the past behavior experience would be changed to the customers behavior clicks on every items. Now there would be no need to predict the items before customers demand the item. The company should not only predict or rely on the total numbers of clicks but they should also conduct the surveys from the customers either who order the items but to those customers too who visited the website and ordered the items. Not only that the people who visited their website should.
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A Exhibit 1 Determination of Range of demand for an item New Items A B C D Actual Demand to Forecast Demand ratios year wise (A/F) 1985 0.3 0.1 1.8 1.5 1986 0.2 0.5 1.5 1.8 1987 0.4 1.8 0.8 1.9 1988 1.2 0.6 0.4 0.8 1989 0.6 0.4 0.5 0.7 1990 0.8 0.5 0.9 0.2
Frequency Distribution 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Total
1 2 1 3 3 2 1 3 1 0 0 1 0 0 2 0 0 3 23
Probability Distribution 0.043478 0.086957 0.043478 0.130435 0.130435 0.086957 0.043478 0.130435 0.043478 0 0 0.043478 0 0 0.086957 0 0 0.130435 1 56% of errors remains between 0.6 to 1.8 so it is assumed with 56% probability that every new items will have actual demand between 600 to 1800 units if frozen forecast is 1000 units i.e. 1000 x 0.6=600, 1000 x 1.8 = 1800
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A Exhibit 2:
Determination of commitment Quantity: Selling Price = 30 Discounted Price = 10 Cost = 18 Profit = Price ± Cost Profit = 30-18=12 Loss = Cost ± Discounted Price Loss = 18-10= 8 Fractile = Profit Profit Loss
Fractile = 12/(12+8) = 0.6 Suppose forecast error against 0.6 is 1.16 and frozen forecast is 1000 units then commitment quantity will be 1000 x 1.16 = 1165units
Actual Demand 840 990 780 Forecast Demand 875 850 600
Item Eggshell Azur Heather
A/F
Rank Rank/Total 0.96 1 0.333333333 1.164705882 2 0.666666667 1.3 3 1
WAC L.L. Bean, inc., a catalog business is unable to mach demand with supply and bearing additional cost of $11 million because of lack of stock and $10 million because of having wrong inventory. These problems are occurred because they are not forecasting exactly what are their customers demanding and how much quantity is demanded. L.L. bean usually place order to their vendor twelve or more weeks before when their customer receive their catalog, it is because the lead time for the production is eight to twelve weeks and it is unable to change the commitment after that. Pattern of early season demand is forecasted by deliveries against a commitment in order to make prediction about end season demand. So it is a challenging situation to forecast demand in this situation. The forecasting problems are very important for L.L. Bean and it is very important for them to find an immediate solution of these problems because there marketing philosophy is " sell good merchandise at a reasonable profit, treat your customers like human being they'll always come back for more." so their customers are very important for them and if they are out of stock, they might lose their customers and if they sell their excess stock at discount price while liquidating, then it would be totally again their philosophy of "sell good merchandise at a reasonable profit." L.L. Bean's marketing approach is direct marketing because they access their current and potential customers by sending them their catalog and then their customers order. Thus, they create demand. This approach had been successful because they reached to their customers directly.
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
Although the demand of a customer is very hard to predict but there are also some prominent mistakes observed related to demand forecasting of L.L Bean which might be cause of the problems. Their product line is like a hierarchy which is divided into merchandise groups, demand centers, item sequences and items and they forecast their demand at item level and in demand forecasting there are two major steps, First they forecast the "frozen forecast" by book in this process, 1st they arrange the items and rank them according to expect dollar sales, after that they assign dollar to each item according to its rank then they use role of thumb to check that if it feels good or if it makes sense. after that they make adjustment if any new item is added to analyze if the item is creating incremental demand or stealing demand of other items. Thus, forecasting is done book by book and item by item and finally frozen and handover the inventory managers. if we look at the whole process of making "frozen forecast" we will come to know that they commonly use the rule of thumb. This was acceptable in the early ages but it is unacceptable in the era of improving technology where the mean of advertisement is changed because of internet and global competition has been started and customer's demand has become highly unpredictable especially, if Scott Sklar, a buyer of men's shirt for L.L. Bean, want to add a new item he would have to observe the trend and gather the information of buyer's behavior related to that "new" item rather than self prediction. with the result of that forecasting, they always observed that the sum of item forecasted remains higher than the dollar targeted to that book so they were uncertain about which item forecasting should be reduced or reduce the whole in some percentage so this was one of the result of under stocking the inventory.
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
In the second step they determine the range of quantity demanded for any (new) item of a product. For this, they take historical errors of each product. value of historical error is obtained by dividing actual demand with forecasted demand for each product. then they compile
frequency distribution of the historical errors then they use that frequency distribution as probability distribution to forecast the range of demand of such item. (see Exhibit 1) They always use A/F ratio for inventory decisions because they want to determine the range of forecast demand for current year and number of units added to stock they do this by multiplying frozen forecast with the A/F ratio. A/F ratio also guides to increase or decrease the frozen forecast for each item by comparing with past data. for instant, if previous forecasted demand exceeds previous actual demand then A/F ratio will decrease current year forecasted demand for that item and vice versa. e.g. previous year forecasted demand is 900 units actual demand is 800 units then A/F is 800/900= 0.88 and if frozen forecast is 1000 then 1000X0.88= 888units. If we look at the process of forecasting range of actual demand for a new, or any other item, item's actual demand would be same like previous pattern of demand but it is not compulsory as Rol Fessenden expressed concern that the there as huge dispersion of errors in "never out" items the dispersion vary at the level of item level like distribution of women's sweeter have more dispersion that than men footwear. it should not be consider that every "never out" item will have same actual demand like all other never out items. The next step is to determine the number of units added to the stock. for this, they consider item's selling price, cost of item, and salvage value or discount price of item and balance the individual item's contribution margin. They calculate the fractile using these costs in order to determine the
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A
number of units added to inventory for this they match the fractile ratio with demand probability distribution according to the rank of item it is same as the newsvendor model (see exhibit 2) Mark Fasold, vice president is concerned that number of items which are purchased are usually exceed the number of items because of large dispersion of errors and the errors are used to forecast actual demand so this dispersion disturbs the average value and demand is more forecasted moreover inventory manager makes the policy to gain high profit rather than avoiding losses. In other words he thinks that it is acceptable to bear fewer losses for gaining high profit. The inventory manager should try to balance the actual and forecasted demand rather than adopting that approach. Solution and Recommendations: To solve the company's major problem of matching demand with supply, inventory manager should make forecasting individually of any new item rather than considering "new" or "never out." Because customer's buying behave vary from item to item rather than time of holding that product for the business. For this, they should make the survey to analyze the buying behavior of customers related to that new item. They should do the regression analysis while calculating the incremental demand rather than self prediction and while determining frozen forecast, they should try to consult more and more past data. If the past data is not enough then they should try to gather the data related to behaviors of customer about the items. The company only consults the past data; they can also conduct survey to analyze baying behavior while forecasting frozen forecasts.
Effects of internet: As the competitors of L.L Bean are using internet for the prosperity of their business and they have the salient point of difference among L.L Bean. They should also upgrade their business to the internet. First
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A of all they should contact the service provider to obtain the domain name and register their website. The website should be handled with the networking expert. The website should match the business nature and type. The well designed website would attract more customers. The expenses would be increased but the possibility of revenue would be increased too. It would enable L.L Bean to match the competitors in the open market in the same business. There should be very positive results for upgrading business to internet. First of all the online orders should be made by the customers. The variety could be easily seen in detail and the chances of orders should be increased. The likings and the disliking of the customers could be seen at the spot because of their extended zoom of each product. The cost of opening the new branch or outlet would also be saved. This would be the plus point of the company to reduce the cost and finding the customers at the same place within no time and less cost.
Recommendations related internet era: The experts should add a feature which can calculate the total numbers of clicks hit to every item by the customers. This would enable the management to know the top preferred items and the least preferred items liked or disliked by the customers. The management would be able to focus on those items more which are more likely liked by the customers. The problem which was already faced by the company of wrong predicted items with respect to the past behavior experience would be changed to the customers behavior clicks on every items. Now there would be no need to predict the items before customers demand the item. The company should not only predict or rely on the total numbers of clicks but they should also conduct the surveys from the customers either who order the items but to those customers too who visited the website and ordered the items. Not only that the people who visited their website should.
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A Exhibit 1 Determination of Range of demand for an item New Items A B C D Actual Demand to Forecast Demand ratios year wise (A/F) 1985 0.3 0.1 1.8 1.5 1986 0.2 0.5 1.5 1.8 1987 0.4 1.8 0.8 1.9 1988 1.2 0.6 0.4 0.8 1989 0.6 0.4 0.5 0.7 1990 0.8 0.5 0.9 0.2
Frequency Distribution 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Total
1 2 1 3 3 2 1 3 1 0 0 1 0 0 2 0 0 3 23
Probability Distribution 0.043478 0.086957 0.043478 0.130435 0.130435 0.086957 0.043478 0.130435 0.043478 0 0 0.043478 0 0 0.086957 0 0 0.130435 1 56% of errors remains between 0.6 to 1.8 so it is assumed with 56% probability that every new items will have actual demand between 600 to 1800 units if frozen forecast is 1000 units i.e. 1000 x 0.6=600, 1000 x 1.8 = 1800
10123027, 10123022, 10123006, 10123035, M.B.A Sec. A Exhibit 2:
Determination of commitment Quantity: Selling Price = 30 Discounted Price = 10 Cost = 18 Profit = Price ± Cost Profit = 30-18=12 Loss = Cost ± Discounted Price Loss = 18-10= 8 Fractile = Profit Profit Loss
Fractile = 12/(12+8) = 0.6 Suppose forecast error against 0.6 is 1.16 and frozen forecast is 1000 units then commitment quantity will be 1000 x 1.16 = 1165units
Actual Demand 840 990 780 Forecast Demand 875 850 600
Item Eggshell Azur Heather
A/F
Rank Rank/Total 0.96 1 0.333333333 1.164705882 2 0.666666667 1.3 3 1
