Who Is Really Master Scheduling Your Company?

by Dave Biggs

"Fred, we need 50 of these and 50 of these and 50 of those by the end of the week." Fred never even looks up. He just offers his standard reply, "I'll see what I can do." Fred is the supervisor of work Cell 1 and he is not Customer Service's favorite person. Late delivery to customers is hurting the company's reputation.

And it's mutual. Fred doesn't like Customer Service, either. They're always making promises without even understanding what it takes to get things done. Manufacturing is constantly in trouble because inventory is going up. How can inventory go up and customer service go down? Let's begin with the situation in Fred's work cell.

The economy has been good and orders have been increasing in spite of the delivery problems. The company went through the painful process of downsizing and is trying to stay lean and mean. Instead of hiring new people, Fred has his crew putting in the maximum amount of overtime to try to keep up with the order rate.

In the simplified example, capacity planning for Fred is easy because there are only three part numbers and all of them require about the same amount of time. Fred fabricates parts that are used in the final assembly of the company's three main product lines. Figure 1 shows the bills of material for the products. Fred's work cell number is 1 and he fabricates the 1A's for Product A, 1B's for Product B, and the 1C's for Product C.

The 2A's, 2B's, and 3B's come from work cell 2 and the 3A's, 3B's, and 3C's, come from an outside supplier. When the annual planning was done, the order rate for the combination of Products A, B and C was projected to be a total of 100 units per week. So when the smoke cleared, Fred's head count budget was set at the amount required to build 100 units.

Demand variation has made it difficult to forecast the actual mix of customer orders for A's, B's, and C's. To keep the bases covered, the forecast has been inflated. This is where the confusion begins. The forecast is asking for 50 of Product A, 50 of Product B, and 50 of Product C to be produced each week. The confusion is further aggravated because we have not been able to ship all customer orders on time and requirements for 25 each of products A, B, and C have accumulated in the late bucket.

The combination of forecast and accumulated orders in the late bucket is scheduling Fred's work center to do 75 each of items 1A, 1B, and 1C in the first week. The problem is, Fred has a budgeted head count that can do 100 units. His crew is putting in heavy overtime, which accomplishes another 50 units a week. The math is against work center 1. They are being asked to complete 225 units, but they only have the resources to accomplish 150.

What would you do in Fred's situation? Well, Fred has to decide which 150 units are going to get worked on. It's not just Fred. The supervisor of work cell 2 has to decide which of the 2A's, 2B's, and 2C's is going to get done, and the outside supplier who has the same annual planning process and budget constraints has to decide which of the 3A's, 3B's, and 3C's is going to get done. Figure 2 shows the worst case when they all decide to do something different!

In this example, Fred decided to build the parts to support Products A and B. But the supplier decided not to support Product A and work cell 2 decided not to support Product B, so no product could be built. Unfortunately, inventory went up because Fred's parts either went to a storeroom or was WIP (work in process) inventory.

In reality, the situation is never this bad. Expeditors begin to coordinate the situation by communicating "hot customer needs" to purchasing so the supplier will do at least a few of item 3A. Fred gets expedited to do at least a few 1C's and work cell 2 gets encouragement to do at least a few 2B's. The actual performance ends up somewhere between the worst situation where none of the parts can go into finished items and the best situation where all the parts can go into finished items.

When the master schedule is higher than the capabilities of the work centers and suppliers, individual work units determine the real schedule by picking and choosing what to do. During the week, expeditors modify the default schedule with hot order information. The result is poorly coordinated efforts that are very inefficient. Inventory goes up and customer service goes down.

Seems like the old rock and a hard place. We can't perfectly forecast customer demand and we are harmed financially if we overstaff just to cover demand variation. And it's hard to "just say no" when the customer wants delivery next Friday!