Demand Forecasting

Go back to Tutorial

There are some important differences in the way inventory requirements are determined that are related to the type of demand for the products in question.

Demand Types

  • Independent demand occurs where the demand for one particular product is not related to the demand for any other product.
  • Dependent demand occurs where the demand for a particular product is directly related to another product.

System of Ddemand Rrequirements

  • A push system is the more traditional approach where inventory replenishment is used to anticipate future demand requirements. A push approach to inventory planning is usually based on a set plan that is predetermined according to certain rules of inventory reordering. This approach is a proactive one in the sense that it is planned on the basis of estimated, or forecast, demand for products from customers.
  • A pull system is where the actual demand for a product is used to ‘pull’ the product through the system. The pull approach is a reactive one where the emphasis is on responding directly to actual customer demand, which pulls the required product through the system. The idea of a pull system is that it can react very quickly to sudden changes in demand.

Lead-Time Gap

The total time it takes to complete the manufacture and supply of a product is often known as the logistics lead time. Customers are generally prepared to wait for a limited period of time before an order is delivered. This is the customer’s order cycle time. The difference between the logistics lead time and the customer’s order cycle time is often known as the lead-time gap.

It is the existence of this lead-time gap that necessitates inventory being held. The extent of the lead-time gap, measured in length of time, determines how much inventory must be held. The greater the lead-time gap, the greater the amount of inventory that must be held to satisfy customer requirements. Thus, the more this gap can be reduced, the less inventory will be required.

Demand Forecasting Methods

Different methods of demand forecasting are used to try to estimate what the future requirements for a product or SKU might be so that it is possible to meet customer demand as closely as possible. Forecasting, thus, helps the inventory holding decision process to find answers to questions about what to stock, how much to stock and what facilities are required.

There are several different approaches that can be used for forecasting. These are

  • Judgmental methods– subjective assessments based on the opinions of experts, such as suppliers, purchasing, sales and marketing personnel, and customers. These methods are used when historic demand data are very limited or for new products. They include brainstorming, scenario planning and Delphi studies.
  • Causal methods– used where the demand for a product is dependent on a number of other factors. The main method used is regression analysis, where a line of ‘best fit’ is statistically derived to identify any correlation of the product demand with other key factors.
  • Projective methods– these forecasting techniques use historic demand data to identify any trends in demand and project these into the future. They take no direct account of future events that may affect the level of demand.

Guidelines for effective demand forecasting are

  • Ensure from the outset that there is a clear plan for identifying and using the most appropriate factors and methods of forecasting. Understand the key characteristics of the products in question and the data that are available. Consider the different quantitative and qualitative methods that can be used and select those that are relevant.
  • Take care to review the base data for accuracy and anomalies. Poor data that are analyzed will produce poor and worthless results. Where necessary, ‘clean’ the data and take out any abnormalities.
  • A typical range of company products can and do display very different characteristics. Thus, it is usually necessary to identify key differences at the outset and group products with similar characteristics together.
  • Use statistical techniques to aid the understanding of output and results (standard deviation, mean absolute deviation, etc.). There may be a number of relevant issues that can impact on the interpretation of results: the size of the sample, the extent of the time periods available.
  • Any forecasting system that is adopted needs to be carefully controlled and monitored because changes occur regularly: popular products go out of fashion and technical products become obsolete.


Inventory Performance Measurement

Inventory performance management or supply chain management has emerged as an important business activity these days. The efficiency of the supply chain can be measured based on the size of the inventory investment in the supply chain. The inventory investment is measured relative to the total cost of the goods that are provided through the supply chain.

Two common measures to evaluate inventory efficiency are:

  • Inventory Turnover
  • Weeks-of-Supply

These essentially measure the same thing and mathematically are the inverse of one another.

Inventory Turnover = Cost of goods sold / Average aggregate inventory value

Here, cost of goods sold is the annual cost for a company to produce the goods or services provided to customers; it is sometimes referred to as the cost of revenue. This does not include the selling and administrative expenses o the company. The average aggregate inventory value is the total value of all items held in inventory for the organization value at cost. It includes the raw material, work-in-process, finished goods, and distribution inventory considered owned by the company.


Good inventory turnover values vary by industry and the type of products being handled. At one extreme, a grocery store chain may turn inventory over 100 times per year. Values of six to seven are more typical.

In many situations, particularly when distribution inventory is dominant, Weeks-of-Supply is the preferred measure. This is a measure of how many weeks’ worth of inventory is in the system at a particular point of time.


When company financial reports cite inventory turnover and weeks of supply, we can assume that the measures are being calculated organization wide. In some very low inventory operations, days or even hours are a better unit of time for measuring supply.

Selective Inventory Management

One of the major operating difficulties in the scientific inventory control is an extremely large variety of items stocked by various organizations. These may vary from 10,000 to 100,000 different types of stocked items and it is neither feasible nor desirable to apply rigorous scientific principles of inventory control in all these items. Such an indiscriminate approach may make cost of inventory control more than its benefits and therefore may prove to be counter-productive. Therefore, inventory control has to be exercised selectively. Depending upon the value, criticality and usage frequency of an item we may have to decide on an appropriate type of inventory policy. The selective inventory management thus plays a crucial role so that we can put our limited control efforts more judiciously to the more significant group of items. In selective management we group items in few discrete categories depending upon value; criticality and usage frequency. Such analyses are popularly known as ABC, VED and FSN Analysis respectively. This type of grouping may well form the starting point in introducing scientific inventory management in an organization.

ABC Analysis

This is based on a very universal Pareto’s Law that in any large number we have `significant few’ and `insignificant many’. For example, only 20% of the items may be accounting for the 80% of the total material cost annually. These are the significant few which require utmost attention.

To prepare an ABC type curve we may follow the following simple procedure:


  • Arrange items in the descending order of the annual usage value. Annual usage value = Annual demand x Unit price.
  • Identify cut off points on the curve when there is a perceptible sudden change o1 slope or alternatively find cut off points at top 10% next 20% or so but do not interpret these too literally- rather as a general indicator.


A very simple empirical way to classify items may be adopted as follows:


Average annual usage value X= Total material cost per year

Total number of items

A-Class items ≤ 6X

C-Class items ≥ 0.5X

In between we have B-class items.


Once the items are grouped into A, B and C category, we can adopt different degree ‘ of seriousness in our inventory control efforts. A class items require almost continuous and rigorous control. Whereas B-class items may have relaxed control and C-class items may be procured using simple rules of thumb, as usual.

VED Analysis

This analysis attempts to classify items into three categories depending upon the consequences of material stock out when demanded. As stated earlier, the cost of shortage may vary depending upon the seriousness of such a situation. Accordingly, the items are classified into V(Vital), E(Essential) and D(Desirable) categories. Vital items are the most critical having extremely high opportunity cost of shortage and must be available in stock when demanded. Essential items are quite critical with substantial cost associated with shortage and should be available in stock by and large. Desirable group of items do not have very serious consequences if not available when demanded but can be stocked items.

FSN Analysis

Not all items are required with the same frequency. Some materials are quite regularly required, yet some others are required very occasionally and some materials may have become obsolete and might not have been demanded for years together. FSN analysis groups them into three categories as Fast-moving, Slow-moving and Non-moving (dead stock) respectively. Inventory policies and models for the three categories have to be different. Most inventory models in literature are valid for the fast-moving items exhibiting a regular movement (consumption) pattern.

Activity A

  • Collect consumption data for 100 different items for an organization and classify these into an ABC framework following the procedure described.
  • List these items in a two-way classification ABC and VED and identify the number of items belonging to each of these 9 distinct groups.

Aggregate Inventory Planning

Aggregate planning is a marketing activity that does an aggregate plan for the production process, in advance of 6 to 18 months, to give an idea to management as to what quantity of materials and other resources are to be procured and when, so that the total cost of operations of the organization is kept to the minimum over that period.


The quantity of outsourcing, subcontracting of items, overtime of labor, numbers to be hired and fired in each period and the amount of inventory to be held in stock and to be backlogged for each period are decided. All of these activities are done within the framework of the company ethics, policies, and long-term commitment to the society, community and the country of operation.


Aggregate planning is the process of developing, analyzing, and maintaining a preliminary, approximate schedule of the overall operations of an organization. The aggregate plan generally contains targeted sales forecasts, production levels, inventory levels, and customer backlogs. This schedule is intended to satisfy the demand forecast at a minimum cost. Properly done, aggregate planning should minimize the effects of shortsighted, day-to-day scheduling, in which small amounts of material may be ordered one week, with an accompanying layoff of workers, followed by ordering larger amounts and rehiring workers the next week. This longer-term perspective on resource use can help minimize short-term requirements changes with a resulting cost savings.


In simple terms, aggregate planning is an attempt to balance capacity and demand in such a way that costs are minimized. The term “aggregate” is used because planning at this level includes all resources “in the aggregate;” for example, as a product line or family. Aggregate resources could be total number of workers, hours of machine time, or tons of raw materials. Aggregate units of output could include gallons, feet, pounds of output, as well as aggregate units appearing in service industries such as hours of service delivered, number of patients seen, etc.

Aggregate planning does not distinguish among sizes, colors, features, and so forth. For example, with automobile manufacturing, aggregate planning would consider the total number of cars planned for not the individual models, colors, or options. When units of aggregation are difficult to determine (for example, when the variation in output is extreme) equivalent units are usually determined. These equivalent units could be based on value, cost, worker hours, or some similar measure.


Aggregate planning is considered to be intermediate-term (as opposed to long- or short-term) in nature. Hence, most aggregate plans cover a period of three to 18 months. Aggregate plans serve as a foundation for future short-range type planning, such as production scheduling, sequencing, and loading. The master production schedule (MPS) used in material requirements planning (MRP) has been described as the aggregate plan “disaggregated.”


Steps taken to produce an aggregate plan begin with the determination of demand and the determination of current capacity. Capacity is expressed as total number of units per time period that can be produced (this requires that an average number of units be computed since the total may include a product mix utilizing distinctly different production times). Demand is expressed as total number of units needed. If the two are not in balance (equal), the firm must decide whether to increase or decrease capacity to meet demand or increase or decrease demand to meet capacity. In order to accomplish this, a number of options are available.


Options for situations in which demand needs to be increased in order to match capacity include:


  • Varying pricing to increase demand in periods when demand is less than peak. For example, matinee prices for movie theaters, off-season rates for hotels, weekend rates for telephone service, and pricing for items that experience seasonal demand.
  • Advertising, direct marketing, and other forms of promotion are used to shift demand.
  • Back ordering. By postponing delivery on current orders demand is shifted to period when capacity is not fully utilized. This is really just a form of smoothing demand. Service industries are able to smooth demand by taking reservations or by making appointments in an attempt to avoid walk-in customers. Some refer to this as “partitioning” demand.
  • New demand creation. A new, but complementary demand is created for a product or service. When restaurant customers have to wait, they are frequently diverted into a complementary (but not complimentary) service, the bar. Other examples include the addition of video arcades within movie theaters, and the expansion of services at convenience stores.

Options which can be used to increase or decrease capacity to match current demand include:


  • Hire/lay off. By hiring additional workers as needed or by laying off workers not currently required to meet demand, firms can maintain a balance between capacity and demand.
  • By asking or requiring workers to work extra hours a day or an extra day per week, firms can create a temporary increase in capacity without the added expense of hiring additional workers.
  • Part-time or casual labor. By utilizing temporary workers or casual labor (workers who are considered permanent but only work when needed, on an on-call basis, and typically without the benefits given to full-time workers).
  • Finished-goods inventory can be built up in periods of slack demand and then used to fill demand during periods of high demand. In this way no new workers have to be hired, no temporary or casual labor is needed, and no overtime is incurred.
  • Frequently firms choose to allow another manufacturer or service provider to provide the product or service to the subcontracting firm’s customers. By subcontracting work to an alternative source, additional capacity is temporarily obtained.
  • Cross-training. Cross-trained employees may be able to perform tasks in several operations, creating some flexibility when scheduling capacity.
  • Other methods. While varying workforce size and utilization, inventory buildup/backlogging, and subcontracting are well-known alternatives, there are other, more novel ways that find use in industry. Among these options are sharing employees with counter-cyclical companies and attempting to find interesting and meaningful projects for employees to do during slack times.

Aggregate Planning Strategies

There are two pure planning strategies available to the aggregate planner: a level strategy and a chase strategy. Firms may choose to utilize one of the pure strategies in isolation, or they may opt for a strategy that combines the two.

Level StrategyA level strategy seeks to produce an aggregate plan that maintains a steady production rate and/or a steady employment level. In order to satisfy changes in customer demand, the firm must raise or lower inventory levels in anticipation of increased or decreased levels of forecast demands.

A second alternative would be to use a backlog or backorder. A backorder is simply a promise to deliver the product at a later date when it is more readily available, usually when capacity begins to catch up with diminishing demand. In essence, the backorder is a device for moving demand from one period to another, preferably one in which demand is lower, thereby smoothing demand requirements over time.

A level strategy allows a firm to maintain a constant level of output and still meet demand. This is desirable from an employee relations standpoint. Negative results of the level strategy would include the cost of excess inventory, subcontracting or overtime costs, and backorder costs, which typically are the cost of expediting orders and the loss of customer goodwill.

Chase Strategy – A chase strategy implies matching demand and capacity period by period. This could result in a considerable amount of hiring, firing or laying off of employees; insecure and unhappy employees; increased inventory carrying costs; problems with labor unions; and erratic utilization of plant and equipment. It also implies a great deal of flexibility on the firm’s part. The major advantage of a chase strategy is that it allows inventory to be held to the lowest level possible, and for some firms this is a considerable savings. Most firms embracing the just-in-time production concept utilize a chase strategy approach to aggregate planning.

Most firms find it advantageous to utilize a combination of the level and chase strategy. A combination strategy (sometimes called a hybrid or mixed strategy) can be found to better meet organizational goals and policies and achieve lower costs than either of the pure strategies used independently.

Aggregate Planning Techniques

Techniques for aggregate planning range from informal trial-and-error approaches, which usually utilize simple tables or graphs, to more formalized and advanced mathematical techniques. William Stevenson’s textbook Production/Operations Management contains an informal but useful trial-and-error process for aggregate planning presented in outline form. This general procedure consists of the following steps:


  1. Determine demand for each period.
  2. Determine capacity for each period. This capacity should match demand, which means it may require the inclusion of overtime or subcontracting.
  3. Identify company, departmental, or union policies that are pertinent. For example, maintaining a certain safety stock level, maintaining a reasonably stable workforce, backorder policies, overtime policies, inventory level policies, and other less explicit rules such as the nature of employment with the individual industry, the possibility of a bad image, and the loss of goodwill.
  4. Determine unit costs for units produced. These costs typically include the basic production costs (fixed and variable costs as well as direct and indirect labor costs). Also included are the costs associated with making changes in capacity. Inventory holding costs must also be considered, as should storage, insurance, taxes, spoilage, and obsolescence costs. Finally, backorder costs must be computed. While difficult to measure, this generally includes expediting costs, loss of customer goodwill, and revenue loss from cancelled orders.
  5. Develop alternative plans and compute the cost for each.
  6. If satisfactory plans emerge, select the one that best satisfies objectives. Frequently, this is the plan with the least cost. Otherwise, return to step 5.

Certified Inventory and Warehouse Analytics Professional

Go back to Tutorial

Get industry recognized certification – Contact us