It is not a game of chance, although it often seems that many organizations have chosen to "guesstimate" their inventory needs. Others install features to assist with demand planning, but their intentions are often better than their implementations because they do not utilize the capabilities to their full potential.

The bottom line is that planning for future demand is truly not a game--it is a skill that relies on data collection and analysis. You have to know the issues, the drivers behind the issues, and the solutions for the best possible forecasting and inventory management.

Inventory Issues: Numerous inventory obstacles occur for manufacturers, distributors, and retailers in both the financial and customer service areas. While they are commonly associated with newly introduced stock keeping units (SKUs), unfortunately, they are also very common with long-standing SKUs as well. The most problematic situations that occur, beginning with the most critical, are:

1. Out-of-stock situations that reduce revenue and cause low order fill rates that frustrate customers due to the less-than-ideal capability to fill customer orders completely and promptly.
2. Long, unknown or unreliable lead times for replenishment.
3. Inventories with discontinued and obsolete SKUs that have no demand at all.
4. Overstock of slow-moving, active SKUs that produce low gross margin return on the investment and the associated low inventory turnover.
5. Non-optimum inventory deployment, which means there is an appropriate total inventory level but it is not positioned in sync with the demand, requiring both transportation cost and transit time to serve customers.

Overstock and Out-of-Stock Drivers: Both overstock and out-of-stock situations are driven by related factors. The primary drivers of inventory issues are forecast accuracy, safety stock policies, lead times, and stocking policies.

With forecast accuracy, over-forecasting what the demand will be creates overstock issues, low turnover, and in the worst cases, liquidation along with the associated write-off. On other hand, under-forecasting demand creates out-of-stock situations and in some cases, an inordinate amount of extra expenses due to attempts to address the issue. Even when forecasts are fairly accurate, inadequate safety stock inventories that are unable to cover short-term demand peaks and valleys have the potential to create out-of-stock situations.

Unpredictable lead times may also cause havoc and reduce forecast accuracy by either being too early or too late. If purchases are placed early, inventory is increased to minimize the possibility of running out of product. This decreases the turnover rate and creates a surplus of inventory. However, if purchase orders are not placed early enough, out-of-stock situations are prone to occur.

Stocking also plays a large role in creating inventory issues. If SKUs are stocked in too many locations for efficiency purposes, transportation costs and customer delivery times are reduced, but inventories are higher, which raises the risk of future liquidation. Contrarily, if SKUs are stocked in too few locations, transportation costs are increased and customer delivery times are extended.

Forecasting and Inventory Solutions: Many firms ineffectively forecast demand, while a remarkable number do not forecast demand at all--not even as simplistically as the past three months of inventory demand divided by three (which is not entirely inaccurate and not recommended).

Furthermore, for those that forecast demand, many do so in isolation with little or no discussion with either sales personnel or customers. Open communications with customers, a sales and operations planning (S&OP) communication process, a good statistical forecasting tool, and a solid instrument for measuring and improving accuracy are all necessary. For demand streams that include one-time events--such as sales and promotions--a strong forecast editing capability is an additional necessity because the history can be so misleading.

Additionally, solid customer service-or gross margin-based safety stock policies are essential. Equally important is a good understanding of management priorities for in-stock positions across the product line, as well as an even better understanding of the relationship between inventory working capital and increased (or lost) revenue. Many firms that purchase software for inventory management or replenishment buying have defined parameters that are unclear or inappropriately set. This undermines both the software capabilities along with the resulting inventory performance.

Simple yet straightforward communication is necessary to reconcile purchaser expectations with vendor realities. Then, as in the case of forecasting, a tool that focuses on mutual understanding can best measure, report, reduce initially variability, and later, measure and report total elapsed time.

Cultivate well-trained personnel to manage inventories, including measuring, reporting, and continually improving customer service and financial performance. Without demand planning strategy in place, you can be sure that you are either over-investing in your inventory, damaging your customer service, or both.

About the Author: Ralph Cox is a principal at Tompkins Associates. The firm designs and integrates global end-to-end solutions for companies that embrace supply chain excellence.
Industry Week