Axe Ski Shop has five large retail stores across three westernstates. They operate a central warehouse that receives theirclothing, ski equipment, and snowboard equipment from variouswholesalers. (The central warehouse also has offices for the Axesystem employees.) Each retail location currently sends weeklyorders to the central warehouse and the warehouse then sends atruck to each location with their weekly orders. This past seasonwas a disaster. Several of the retail locations regularly ran shorton ski and snowboard boots, one of the retail locations had excessclothing inventory all winter long, and customers’ special requeststo the warehouse either were ignored or were received late. In somecases the items would show up on the weekly delivery but were wrongsizes. Overall, customer satisfaction dropped and online reviews oftheir stores has reached an all-time low.

To fix these issues Axe’s central management has decided toinstall a company-wide communication system for their inventory soall five retail locations and the central warehouse can seeinventory levels by SKU at all locations. Orders to the warehousewill be automatically generated based on inventory levels whichwill alleviate several of the issues described above. This newsystem will also let the retail locations request items shippedfrom store to store as needed to meet customers’ specialrequests.

The communication system will be rolled out incrementally toeach location over the spring/summer months (shown in Table 1),starting with the warehouse, so that the system is fullyfunctioning by August. Axe wants to purchase and install theoptimum number of servers required to run the system but isn’t surewhether they need to install all servers in March or if they caninstall the servers incrementally as well. The number of employeesat each location dictates when server capacity is needed, shown inTable 2. Ultimately, the system needs to handle all companyemployees. There is variation in server capacity, shown in table 3below.

Additional cost and budget information includes a maximum budgetof $18,000 total for the months of March and April. A discountduring these same months is available on the workstation servers(10%) and the full rack servers (25%). Any servers purchased duringMay-July will not receive a discount.

1. Set up and solve five different optimization spreadsheet modelsto determine how many servers to purchase during each month,March-July. Answer the following questions:
   1. How many of each server is purchased and when?
   2. What is the total capacity, in number of employees, of thesystem?
   3. What is the total cost if this purchase schedule isfollowed?
2. Set up a sixth optimization spreadsheet model that considersthe five months as a whole.
   1. How many of each server is purchased and when?
   2. What is the total capacity, in number of employees, of thesystem?
   3. What is the total cost if this purchase schedule isfollowed?
   4. How much savings is realized by solving this for the whole fivemonths instead of month by month?
3. If you could relax one of the constraints, could you save Axeeven more? Explain the change you would suggest and the resultingsolution.

Table 1

March	April	May	June	July
Central Warehouse	Denver	Salt Lake	Colorado Springs	Bozeman Boise

Table 2

Location	# of Employees
Central Warehouse	50
Denver	200
Salt Lake	180
Colorado Springs	75
Bozeman	35
Boise	30

Table 3

Type of Server	Number of Employees Supported	Cost of Server
MiniDesktop	Up to 30	$3000
Desktop	Up to 80	$5500
Workstation	Up to 200	$12000
Full Rack	Up to 1000	$25000