Algunas confesiones han revelado datos sobre la magnitud de los crímenes, pero aún falta mucho por esclarecer

Demand management is the function of recognizing all demands for products and services in order to support the customers in the marketplace. This demand management process also includes doing what is required to help make the demand happen and prioritizing demand when supply is lacking. Conversely, demand management is also concerned about recognizing when demand is substantially unusual (either positive or negative) and how plans need to be changed to react. Demand management facilitates the planning and utilization of resources for profitable business results. Overall demand management maintains an intense focus on the customer. Demand management can be

broken into two levels of control: demand planning, and demand control and execution. Demand planning includes forecasting, customer linkage, new product planning and inventory and capacity strategy. Demand control and execution includes customer order fulfillment, forecast consumption, abnormal demand management, and safety stock or buffer management.

Demand planning is directly dependent on the quality of the demand information. The process must be defined and enabled so that the best, most timely demand information is available. This process includes a forecast review including accountability. To facilitate this process many companies are establishing the position of demand manager or demand coordinator in order to facilitate and coordinate the development of the demand plans for the business. There should be a formal management demand review of the monthly updated demand plan/forecast as part of the Sales and Operations Planning process. One of the most difficult parts of demand management is distinguishing between small changes in demand patterns that are really a signal of significant shifts in the market or one-time demand spikes that are unusual and not recurring. Reacting to every change in demand would cause terrible disruption in the enterprise and throughout the supply chain. Effective demand management is driven by a clear understanding of the underlying causes of demand as well as the contributors for variability from expected demand. A closed loop sales and operations planning system is a key success factor in effective demand management. This is covered in more depth in Chapter 3.

Forecast

Usually the demand planning process begins with a forecast. Businesses do not work well without a reasonable forecast. The forecast does not need to be 100% accurate. Unless the total lead-time of the supply chain is less than the customer’s response time expectation, a forecast is necessary in order to drive any business. Good usable forecasts result from hard work, timely reviews, and updates, by the appropriate people using customer input, judgment, and software support. The answer to good forecasts is not a more expensive forecasting system. More importantly, the forecast and management of the business need to be at the same level of aggregation. The strategic vision and business plan should be supported by the forecast. The forecast also helps provide a target for sales. This, in turn, through the material and capacity planning process, defines the required resources (equipment, people, materials, and services) to support the plan. The integrated plan provides visibility that is required to reach consensus on the support needed to service the customer at the desired level. The forecast is a key driver of the business planning process.

Demand patterns can vary dramatically and it is very important they are well understood. These patterns must be analyzed and measured to improve our ability to understand how to deal with them. This understanding also provides an idea on how accurate our forecasts can be. A basic principle of forecasting is that a forecast is only as good as the assumptions behind it. Sharing the assumptions behind a forecast leads to a shared understanding of the business. When forecasts are wrong it is very likely due to the assumptions that were behind them.

Balance Supply and Demand

Demand management attempts to balance supply and demand. Demand can come from forecasts, customer orders, distribution, spare parts, safety stock, interdivisional requirements, and new products. Supply can come from material in inventory or expected on factory orders, capacity (labor or machine), and directly from suppliers. When demand exceeds supply there must be an effective demand prioritization process in effect until the supply can be increased.

For the generic enterprise the order entry and promising process is shown in Figure 8.1.

In reality the order entry and promising process must address the following three questions:

1. What is the best business practice for the company? 2. Who needs to be involved?

3. What mechanics are required?

Proper forecast consumption mechanics can be crucial to having demand stability. The first attempt at forecast consumption logic is a simple

Figure 8.1 The order entry and promising process. (From Richard C. Ling, Inc., copyright 2003. With permission.)

blending approach. If the monthly forecast is 1000 units, the spread would simply be done as 250 units per week or 50 units per day. Although this process is very simple to understand, it does not work very well. Assume 3 days pass and no customer orders are received. What should the forecast be for the fourth day? The options that exist are to drop the forecast by the unconsumed units or carry the unconsumed forecast into the next

time period. Another alternative is to spread the unconsumed forecast evenly across the balance of the month. Now you can understand why software companies have been slow to develop demand management systems. The supply side is much simpler and straightforward.

If forecasts are done in monthly time increments then it probably makes sense to consume the forecast with incoming customer orders over that same time frame. Any forecast left over at the end of the month will be deleted until you know something that tells you to do something different. In any event, all incoming customer orders should be screened for normal/abnormal demand characteristics. Practically speaking this is not possible for every customer order, but it must be done on an case-by-case basis to allow correct forecast consumption. Identifying abnormal customer demand for the individual customer order may include considering the demand source—either customer or channel. If all of a sudden demand for an industrial product is experienced from the retail channel, this demand should set a flag for review. The quantity on the order may also trigger the need for review. This could be the size of the order or the percentage of an item’s forecast that is consumed by one order. Triggers can also be set for cumulative demand variances. Abnormal demand types include:

1. One time demands—This order will never be seen again and should not be included in future forecasts.

2. Ordering pattern change for timing—This could be due to a budget cycle being reset or pricing strategy change. This would impact future forecasts if the new budget cycle or pricing strategy were expected to remain.

3. Time to change the forecast—Abnormal demand may point out where the forecasting process or underlying assumptions are in error.

4. Wrong seasonal timing pattern—This may sometimes be caused by an alternate use being developed for a product such as a waterski board now being used as a snowboard.

5. Spiked demand—This could be absolutely normal if the demand is dependent on inventory in the supply chain. In this case forecasting is not the right answer. Better results would be obtained using Distribution Requirements Planning. See Chapter 6 and Chapter 11 for more details on this tool.

In order to maintain maximum user control, there needs to be the ability to code each abnormal demand with a flag so that the planning system will not consume the forecast. In addition, future forecasts should not include this quantity as part of the mathematical computation. A policy is necessary to state there will be a process and mechanism that helps to identify and deal with abnormal demands. Accountability needs to be established in order entry, customer service, and the demand management functions. Many companies must deal with multiple demand streams for the same item. This item may be sold as an end item and is promised through the use of available to promise (ATP). The same item may also be used as part of another assembly and is controlled by MRP. These demand types must be included in the overall company policy.

Safety Stock

Safety stock is a request for inventory to buffer demand or supply variation to better serve the customer. Safety stock is directly related to Demand Management (covered previously in this chapter) and the overall buffer resource strategy (see Chapter 4). The other buffer strategy is to have safety capacity that is a request for available surge capacity in order to support abnormal demands. The better term for safety stock is strategic inventory or buffer inventory. A question that is commonly asked is, “How much safety stock should you have?” This level is usually statistically calculated to maintain a specific service level. This is covered in more detail later in this chapter and in Chapter 11. Safety stock can also be visually determined by reviewing demand variation. New products are commonly supported with subjective safety stock estimates or estimates based on other similar product introductions. In a supplier partnership, the amount of safety stock may be directly negotiated as part of the procurement contract. The ownership of this safety stock may be the supplier’s or the customer’s. This is covered in more detail in Chapter 6 on Strategic Sourcing.

No matter how the safety stock is determined, the important thing to remember is that the system will treat it like a demand rather than a supply. Therefore safety stock needs to be managed and consumed just like a forecast. If the safety stock is never used, then it is not really needed. The Enterprise Resource Planning (ERP) system should provide visibility of when the safety stock has been used to fill customer demand. Customer order allocations should also allow the use of the quantity that has been identified as safety stock. If safety stock is being used, it must eventually be replenished. This replenishment should occur with a lead-time that makes sense in your environment. Unfortunately, most ERP material planning functions treat safety stock as a requirement in the current time period. This causes a past due expedited order to be created to simply replenish safety stock that was doing exactly what was intended. It is difficult for the planner to distinguish these expedite messages from the messages that directly impact a customer. Having analytics and planning functions sufficiently sophisticated to identify this difference makes the production planning function more real.

When demand increases rapidly and resources are having a difficult time keeping pace, this is a good time to reduce safety stocks and possibly even prioritize customers. This prioritization can also get to the point of product allocation if capacity cannot be sufficiently increased in time. Conversely as soon as it is apparent that the demand that was anticipated is not going to happen, the forecast and safety stock should be reduced so material and capacity resources can be freed up to handle real demand. This is not an easy process to accomplish, but it is extremely helpful to the internal manufacturing resources and the overall customer service. Manufacturing can then work on those things that are really demanded by the customer rather than replenishing unneeded safety stocks.