Computco sells high-end computer workstations. The demand for its workstations during a month follows a normal distribution, with a mean of 400 and standard deviation of 100. Each time an order is placed, costs of $600 per order and $1500 per workstation are incurred. Workstations are sold for $2800, and if Computco does not have a workstation in stock, the customer will buy one from a competitor. At the end of each month, a holding cost of $75 per workstation is incurred. Orders are placed at the end of each month, and they arrive at the beginning of the next month. Four ordering policies are under consideration:

■ Policy 1: Place an order for 900 workstations whenever the end-of-month inventory is 50 or less.

■ Policy 2: Place an order for 600 workstations whenever the end-of-month inventory is 200 or less.

■ Policy 3: Place an order for 1000 workstations whenever end-of-month inventory is 400 or less.

■ Policy 4: Place an order for 1200 workstations whenever end-of-month inventory is 500 or less.

Using simulation, run 1000 iterations of an appropriate model to determine which ordering policy maximizes expected profit for a two-year period. To get a more accurate idea of expected profit, you can credit Computco with a salvage value of $1500 for each workstation left at the end of the last month. Assume that 400 workstations are in inventory at the beginning of the first month.