What Is Demand Forecasting?
Picture the holiday season. Kids are ready for a visit from Santa, and parents are stressed out over shopping and finances. Businesses are finalizing their operations for the calendar year and preparing to move into whatever lies ahead.
ABC Inc. manufactures telephone wire. Their accounting and operations time periods run on a calendar year, so the end of the year allows them to wrap up operations before the holiday break and plan for the beginning of a new year. It's time for managers to prepare and submit their department's operational plans to senior management so they can create an organizational operations plan for the new year.
The sales department is stressed out of their minds. Demand for telephone wire was down in 2015, and the general economic data suggests a continuing downturn in construction projects that require telephone wire. Bob, the sales manager, knows that senior management, the board of directors, and stakeholders are hoping for an optimistic sales forecast, but he feels the ice of industry recession creeping up behind him to tackle him.
Demand forecasting is the method of projecting customer demand for a good or service. This process is a continual, where managers use historical data to calculate what they expect the sales demand for a good or service to be. Bob uses information from the company's past and adds it to economic data from the marketplace to see if sales will grow or decline. Bob uses the results of demand forecasting to set goals for the sales department, while trying to keep in line with company goals. Bob will be able to evaluate the results of the sales department next year to determine how his forecast came out.
Bob can use different techniques that are both qualitative and quantitative to determine the growth or decline of sales. Examples of qualitative techniques include:
- Educated guesses
- Prediction market
- Game theory
- Delphi technique
Examples of quantitative techniques include:
- Data mining
- Causal models
- Box-Jenkins models
These examples of demand forecasting techniques are only a short list of the possibilities available to Bob as he practices demand forecasting. This lesson will focus on two additional quantitative techniques that are simple to use and provide an objective, accurate forecast.
Moving Average Formula
A moving average is a technique that calculates the overall trend in a data set. In operations management, the data set is sales volume from historical data of the company. This technique is very useful for forecasting short-term trends. It is simply the average of a select set of time periods. It's called 'moving' because as a new demand number is calculated for an upcoming time period; the oldest number in the set falls off, keeping the time period locked. Let's look at an example of how the sales manager at ABC Inc. will forecast demand using the moving average formula.
The formula is illustrated as follows:
|Moving Average = (n1 + n2 + n3 + ...) / n|
Where n = the number of time periods in the data set. The sum of the first time period and all additional time periods chosen is divided by the number of time periods. Bob decides to create his demand forecast based on a 5-year moving average. This means that he will use the sales volume data from the past 5 years as the data for the calculation.
|Year||Sales Volume ($M)|
|Moving Average n = 5 = (4.6 + 5.3 + 8.1 + 7.8 + 8.3) / 5||6.82|
The sales manager forecasts the sales demand for 2016 to be approximately $6.82 million. To forecast the sales demand for 2017 still using the 5-year moving average, simply drop off 2011 and add 2016.
|Year||Sales Volume ($M)|
Exponential smoothing is a technique that uses a smoothing constant as a predictor of future forecasting. Whenever you use a number in forecasting that is an average, it has been smoothed. This technique takes historical data from previous time periods and applys the calculation for exponential smoothing to forecast future data. In this case, Bob will also apply exponential smoothing to compare against the earlier calculation of a moving average to get a second opinion.
The formula for exponential smoothing is as follows:
- F(t) = forecast for 2016
- F(t-1) = forecast for previous year
- alpha = smoothing constant
- A(t-1) = actual sales from previous year
The smoothing constant is a weight that is applied to the equation based on how much emphasis the company places on the most recent data. The smoothing constant is a number between 0 and 1. A smoothing constant of 0.9 would signal that management places a lot of emphasis on the most previous time period's historical sales data. A smoothing constant of 0.1 would signal that management places very little emphasis on the previous time period. The choice of a smoothing constant is hit or miss and can be modified as more data is available. We will use the chart from above with the historical sales volume to calculate the exponential smoothing forecast for 2016. There is an extra column to include forecasted sales volume.
|Year||Sales Volume ($M)||Forecast|
The calculation can be laid out as follows, using a smoothing constant of 0.5:
|F (2016) = 7.73 + 0.5 (8.3 - 7.73)|
Once the calculations are complete, we find that the forecast for 2016 is equal to 8.02. This calculation is a fairly efficient formula and quite accurate compared to other techniques of demand forecasting.
Demand forecasting is an essential part of a company's projected plans for future time periods. Different techniques can be used, both qualitative and quantitative, and provide differing sets of data to managers as they forecast demand, especially in sales volume. The moving average and exponential smoothing techniques are both fair examples of methods to use to help forecast demand.
To unlock this lesson you must be a Study.com Member.
Create your account
Moving Averages and Exponential Smoothing:
Calculation Problem 1
FunkyTunes has revenue in January of $5000, in February of $6000, in March of $7000, and in April $8000. Forecast the revenue for May using a three-month moving average.
Using the same data, assume the forecast for April was $8200. If FunkyTunes uses a smoothing constant of 0.6, what would be the forecast for May using exponential smoothing>
Calculation Problem 2
We-B-Tools has a sales forecast of 630 tools for November. After November is over, the actual sales amount was 600. How much is the error in the forecast (use the absolute value)? What is the percentage error of the actual sales amount?
Demand is not the only thing that forecasting can predict. There are many things in the economists that economists try to predict. Using the Internet, find three things that economists try to predict on a monthly basis.
1. Under what situations would you consider a moving average to be a better forecast model than exponential smoothing? Why?
2. When do you think a two-month moving average is better than a three-month moving average? What about a four-month moving average? Why?
3. Typically, when one discusses demand, business sales are referenced. However, consider an Internet server. Web browsers all over the world create a demand for information from this server. For businesses, we often think of demand in terms of months and days. For an Internet server, what time frame do you think would be appropriate? Why?
Register to view this lesson
Unlock Your Education
See for yourself why 30 million people use Study.com
Become a Study.com member and start learning now.Become a Member
Already a member? Log InBack