Car 1 is at the attached window and will require minutes to service. Car 2 must wait for car 1 to be serviced minutes the queue + minutes to be serviced = 6 minutes). Car arrived at the attached window minutes after cars 1 and . It must wait 1 minute for car 1 to be serviced and minutes for car to be serviced, a waiting time of 4 minutes. A bank has two drive-in lanes to serve customers: one attached to the bank itself and one on an island. One teller serves bath stations. The bank is interested in determining the average waiting times of customers and has developed a model based on random numbers. The two key factors are the time between successive car arrivals and the time customers wait in line. Assume that the analysis begins with cars just arriving at bath service windows. bath requiring 3 minutes of service time. Car 1 is the attached window attached to the bank unless that window has more cars waiting than the island window. The lone teller will always serve the car that arrived first. If two cars arrive simultaneously, the one at the attached window will be served before the one at the island.

The EOQ minimizes the sum of carrying and setup costs. The EOQ is the amount at which carrying costs are equal to setup costs. Thus, plugging the data into the EOQ formula results in the following:

Thus, if each lot consists of 1.000 units, five production runs per year are needed to meet the 5.000-unit demand. At this level, setup costs will total US $5,000 (5 * $1.000) Carrying costs will also equal US $5,000 ($10 per unit carrying cost * average inventory of 500 units). Accordingly, total costs are minimized at US$10,000.

An economic order quantity EOQ) sensitivity analysis involves varying the holding carrying) costs per unit and/or the order costs to determine how much the changes affect the optimal EOQ.

The simplest way to solve this problem is to use a scatter diagram:

The data points appear to form a straight line with a negative slope. Thus, -1 is the best estimate of the coefficient of correlation.

Sensitivity modeling can be used to determine the outcome of a variety of decisions. A trial and- error method may be adopted, usually in a computer model, to calculate the sensitivity of the solution variability of outcomes) to changes in a variable.