Forecasting Techniques

As part of the inventory system, the ability to forecast future demand patterns is an essential feature.

Some of the possible changes in demand pattern are both of average demand and variation in demand. In addition, within any of these patterns a freak or 'rogue' demand can occur, which may or may not have an identifiable cause.

It is the purpose of forecasting to be able to predict these changes before they occur, to be able to adjust the control parameters within the system.

There are a wide variety of methods for short- and long-term forecasting of demand, varying from guesses/estimates, through simple to extremely sophisticated mathematical techniques.

Inventory systems, in practice, make quantified forecasts and modify or overlay this with known additional information, where appropriate.
It is important to note that the process of forecasting is an attempt to forecast average demand.

Select the Right Inventory Forecasting Model

For wholesalers and distributors of durable goods products, inventory forecasting is especially important as it is the foundation upon which all company plans are built in terms of markets and revenue projections. Management would be a simple matter if business was not in a continual state of motion, the pace of which has quickened in recent years.

Inventory forecasting models are critical elements of the forecasting process, as accuracy can drastically influence business profitability. It is becoming increasingly important and necessary for businesses to predict their future demand in terms of inventory availability, sales assumptions, costs, and profits.

Assessing the actual value of future sales is crucial, as it directly affects future carrying costs and profits, so the prediction of future sales is the logical starting point of all business planning, including inventory purchasing.

Main Inventory Forecasting Model

There are two main inventory forecasting models for enhancing inventory forecasting accuracy to consider:

Quantitative Forecasting

This forecasting approach is a mathematical model based on historical data. It involves using past sales data to predict future demand for goods. Data sets can go back decades or can be run for the last calendar year. However, the more data available, the more accurate picture of historical demand will be obtained. While it may provide a basis for forecasting, demand can be unpredictable, based on variable market conditions or product seasonality. Unexpected peaks in demand can result in stock outages, and quiet periods may result in costly excess stock, which can escalate carrying costs, resulting in diminishing profits.

Qualitative Forecasting

This method is less precise and involves predicting demand based on less-measurable factors, such as market forces, economic demand, and potential demand. Qualitative forecasting methods could be considered an art, mastered by inventory planners over years of practice. Inventory forecasting techniques are inseparable from current stock review and re-order methods, and there are two broad models for inventory monitoring.

Monitoring Inventory and Replenishment

Depending on the industry and the unique businesses inventory turnover ratios, there are two different models for monitoring inventory and replenishment:

Continuous Review

Stock is reviewed on a continuous basis, and a certain amount of stock is re-ordered once levels have dropped below a certain level, known as the re-order point. Having the right re-order point will allow for supplier-variability and should be supported by a safety stock level ‘safety net’ to ensure service level rates are maintained. Companies with fast moving products are highly encouraged to invest into tools that support continuous inventory review.

Periodic Review

Stock levels across all storage points are monitored at set periods, and replacement stock is ordered. This usually results in higher inventory levels, as stock needs to be ordered in sufficient quantities to reach the next review date. This model requires more available cash flow and working capital to maintain and the risk of overstocking and under stocking items is greater. Businesses with slower inventory turnover can manipulate this model successfully with the help of inventory optimization solutions that provide inventory alerts when actions need to be taken to avoid undesirable stocking situations.

Long Term and Short-Term Inventory Forecasting Models

Inventory planners need to evaluate and monitor both long-term and short-term influencers when it comes to demand forecasting. For instance, in a short-run forecast, seasonal demand patterns are of great importance to inventory planners. Seasonality, typically, will influence a period of three months, six months or one year, and these data-points should help guide planners when making inventory top up planning and procurement activities.

Long-term demand forecasting models are helpful when it comes to making larger capital-planning decisions. These models provide information for making major strategic decisions and demand pattern data from long-term data sets can help a company forecast for end of life products and new product introductions to a growing industry.