SBP 10b: Customer Inventory Costs

This principle occurs because customers expect to be served shortly after they demand service (near JIT) and have presented their inputs. Also, waiting can wear down customer-self inputs.

Our approach to managing queuing systems is similar to the way we manage inventories in that we attempt to plan the system so as to minimize total costs. The main difference is which costs dominate. With some manufacturing inventory systems, if we make or purchase too much of an item at one time, we have high costs of holding the items in inventory until we need them, but at least we do not have to set up to make the item for a longer time. Therefore, we trade off inventory holding costs against procurement costs. In other manufacturing systems we trade off the holding costs against the costs of not having enough of an item when it is needed. (Examples are the “perishable inventory model” or the “safety stock” calculation.) In either case, if we run out of the item, we incur back-order costs or a lost sale.

With services, we cannot complete production in advance of demand because we do not yet have the customer inputs (from the Unified Services Theory). Therefore, the queue management issue is how much we should be prepared to produce at the time demand occurs (and customer inputs are received). If the service provider has too little capacity, there will be potentially great costs to the customer (such as inconvenience, lost time, and lost good will with the service provider). You should note that the most significant customer costs are psychological, which can dramatically influence the potential for repurchase and loyalty. Too much capacity lowers the potential costs to the customer, but results in higher costs in terms of idle capacity. Which is preferable? It depends on the estimates of the costs. The fact is, the magnitude of psychological costs of delay to customers are generally much higher than the corresponding monetary costs of delay to manufacturing inventories–which accounts for the significant differences in the acceptable times in inventory between the two systems. Nevertheless, too much service capacity may be good for the customers, but someone has to pay for all of that idle capacity.

Service providers must consider the time costs of customer waiting, and the accompanying psychological costs to the customer. These costs are balanced against the cost of capacity and the costs of reducing negative psychological effects.

When customer waiting costs are lowered by increasing the service capacity, a capacity cost is incurred. The amount of capacity necessary for a corresponding benefit to the customer can be estimated using “Queuing Theory” as was discussed in the Inadvertent JIT Service Business Principle.

Another way to decrease the customer-waiting costs without increasing capacity (the service rate) is through understanding and using the “psychology of queuing.” This psychology captures the idea that costs to customers are not realized according to the actual amount of waiting, but rather according to the amount of waiting perceived by the customer! Therefore, if the perceived wait can be decreased, the waiting cost decreases, even if the actual wait is the same. An oft quoted pioneer of queuing psychology is David Maister. He made the following observations:¹⁾

• Unoccupied waits seem longer than occupied waits.
• Pre-process waits seem longer than in-process waits.
• Anxiety makes waits seem longer.
• Uncertain waits seem longer than waits of a known duration.
• Unexplained waits seem longer than explained waits.
• Unfair waits seem longer than equitable waits.
• The more valuable the service, the longer people will be willing to wait.
• Waiting alone seems longer than waiting with a group.

Maister's points could stand some scientific study, and should be taken with careful thought. (In fact, I am in the middle of a multi-year study of explained waits involving experiments we have conducted around the globe.) Nevertheless, there is great merit in considering the psychological implications of customer inventories.

Post offices are notorious for extended waits at times when it is convenient to visit the post office but not convenient to wait. One day I left my family in the car while I ran into the local post office. To the side of the long line was a screen playing my favorite scene from the Lion King (Akuna Matada). The problem was that the line seemed to move so rapidly that I was unable to see the rest of the scene by the time I was at the front of the line. Had not my family been waiting in the car, I could have gone through the line again to catch the rest of the scene. (Occupied waits seem shorter than unoccupied waits.)

Some amusement parks place signs a ways back in a line telling waiters how many minutes to the front of the line. (Waits of a known duration should seem shorter than uncertain waits.) One time at Universal Studios Florida, my wife and I were waiting in the popular “Back to the Future” ride. The sign said something like “forty five minutes wait from here.” The line sure did not seem like it could be that long of a wait. After waiting perhaps twenty minutes we were already at the front of the line! Although it was only the line we could see from the beginning! There was another line of equal duration on the other side of a partition. After another 20 minutes or so, we started on a line up a ramp to the ride. Sure enough, it was about forty five minutes. It seemed like forty five minutes. Without the sign at the beginning, it might have seemed like more than forty five minutes. The park failed in an opportunity to ease the burden of the line. They had television monitors hanging from the ceiling throughout the line. Every so often, a short segment featuring “Back to the Future” star Christopher Lloyd would come on, which dramatically improved the wait situation. But, most of the time the monitors were blank, which left us to spend our time conversing about the questioned legitimacy of the forecasted forty five minute wait.

Let's come back to the problem of extremely long lines of customers checking in at the gate. I often wonder: “Is such a wait necessary?” “What if I just boarded directly at the gate without checking in?” “What is this so-called boarding pass I got from my travel agent, anyway?” So, I wind up standing in line to prevent a major incident at boarding time because I don't have the right kind of boarding pass. If the airline either (a) put up a sign, or (b) had an employee walk down the line identifying people who do not need to stand in line, the queue would probably be much smaller. In addition, the people who remain in the line would then know it is all for a purpose.

With manufacturing, the customers are not directly involved with inventory costs, but the inventory costs are holding costs incurred to the company.

1. What are the costs to customers in terms of waiting in queues (themselves, their belongings, or their information)? Which of these costs are psychological? Can these costs be estimated in monetary terms?
2. What are the costs of additional capacity?
3. In what ways do some of these costs tend to dominate service process planning decisions? In other words, which costs tend to (or should) drive planning decisions?
4. How might psychological costs of customer waiting be reduced without adding additional capacity?

Develop a plan for managing the psychology of queues. Describe a major customer queue in your service business process (or in an alternate process if necessary). Explain how some of Maister's principles might be applied to decrease the psychological costs of the queue without increasing server capacity. (Or, suppose your own reasonable principles of queuing psychology.) For example, describe an effective way customers could be occupied in a meaningful activity while they are waiting. Or, how could a pre-process wait be turned into an in-process wait by starting customers with the production process while they are waiting? Be creative. (Maister's principles were:

• Unoccupied waits seem longer than occupied waits.
• Pre-process waits seem longer than in-process waits.
• Anxiety makes waits seem longer.
• Uncertain waits seem longer than waits of a known duration.
• Unexplained waits seem longer than explained waits.
• Unfair waits seem longer than equitable waits.
• The more valuable the service, the longer people will be willing to wait.
• Waiting alone seems longer than waiting with a group.)