Home
>
Blogs
>

# Exponential Smoothing

October 19, 2021

There are two main methods in forecasting software: quantitative and qualitative. Quantitative forecasts use data that can be measured, like sales data or business statistics. Qualitative data isn’t measurable, and it’s often derived from expert opinion and experience.

There are multiple approaches that fall under either quantitative or qualitative methods. An accurate forecasting tool takes all approaches into account when producing forecasts.

Exponential smoothing is one approach to quantitative forecasting. This approach uses historical data of demand to produce forecasts. It’s different from the moving averages method, and there are some advantages and disadvantages.

In exponential smoothing, there’s a value associated as a smoothing constant. This number is between 0 and 1. This is done because over time factors change, and forecasts need to reflect those changes. After mathematically determining a smoothing constant, it’s used as the exponent throughout the forecast. The success of exponential smoothing is partially dependent on the experiences and judgment of the demand planner.

## Exponential smoothing formula:

Forecast of period 1 +
α* (Actual Sales for period 1 – Forecast for period 1)

Exponential smoothing is often used in visual aids and presentations. It’s often used in finance and economics settings. If there isn’t a clear pattern, the “smoothing” part of these graphs makes things easier to be explained and understood.

## Benefits of Exponential Smoothing:

1. It’s easy to understand and apply exponential smoothing. The only things needed in this method are the forecast for the latest time period, the actual data for that period, and the smoothing constant. Through trial and error, a demand planner can learn to correctly use the exponential smoothing method and implement it into future forecasts.
2. The more recent the data, the more heavily weighted it is in the exponential smoothing method. While this method often considers a long period of time in forecasts, it’s smart to weight recent data more heavily because that accounts for current consumer demand. For example, January 2018 wouldn’t be weighted as heavily as January 2021, because many elements have changed in those 3 years, and this method tends to project a similar forecast to that of the previous period. This also means that spikes in data don’t have as much of an effect on the forecast as they do in moving averages.
3. Exponential smoothing produces accurate forecasts. Forecasts produced using this method are accurate and reliable and they predict for the next period. The forecast shows projected demand and actual demand. This allows demand planning to be done effectively, therefore resulting in accurate inventory levels.

## Drawbacks of Exponential Smoothing:

1. The ‘smoothing’ part of this method brushes over high and low variations. As the forecast graph shows a smooth line of data, it’s important to note that spikes in data aren’t necessarily represented.
2. Exponential smoothing doesn’t correctly handle trends. This method is best for short term forecasting, for example the next period. It often predicts future patterns to heavily represent current ones, so it’s not as effective in long term forecasting.

A real-life example can be seen in America’s favorite beverage company, Coca-Cola. Coca-Cola uses a simple exponential smoothing forecast. The stock market is unpredictable and hard to understand for many. Coca-Cola’s investors are often unaware of what the value of their stocks will be tomorrow.

Rather than moving averages with equally weighted values, this works better for them. Exponential smoothing allows Coca-Cola to have a more reliable vision of their near future because they can weight things depending on the age of their prices.

At Avercast, we’ve implemented many different methods into our forecasting software. Of course, these methods include exponential smoothing. If you’d like to benefit from the most accurate forecasting tools available, you’ve come to the right place. With 40 years of supply chain experience, we’ve developed the most reliable forecasting software on the market. Our experts have developed 250+ algorithms to ensure our users get the most accurate and reliable predictions possible.

If you’d like to strengthen your demand planning, you need Avercast. Our team would love to connect with you today. Sign up for a free demo or call to learn more!

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.

## Avercast Blogs

There are two main methods in forecasting software: quantitative and qualitative. Quantitative forecasts use data that can be measured, like sales data or business statistics. Qualitative data isn’t measurable, and it’s often derived from expert opinion and experience.

There are multiple approaches that fall under either quantitative or qualitative methods. An accurate forecasting tool takes all approaches into account when producing forecasts.

Exponential smoothing is one approach to quantitative forecasting. This approach uses historical data of demand to produce forecasts. It’s different from the moving averages method, and there are some advantages and disadvantages.

In exponential smoothing, there’s a value associated as a smoothing constant. This number is between 0 and 1. This is done because over time factors change, and forecasts need to reflect those changes. After mathematically determining a smoothing constant, it’s used as the exponent throughout the forecast. The success of exponential smoothing is partially dependent on the experiences and judgment of the demand planner.

## Exponential smoothing formula:

Forecast of period 1 +
α* (Actual Sales for period 1 – Forecast for period 1)

Exponential smoothing is often used in visual aids and presentations. It’s often used in finance and economics settings. If there isn’t a clear pattern, the “smoothing” part of these graphs makes things easier to be explained and understood.

## Benefits of Exponential Smoothing:

1. It’s easy to understand and apply exponential smoothing. The only things needed in this method are the forecast for the latest time period, the actual data for that period, and the smoothing constant. Through trial and error, a demand planner can learn to correctly use the exponential smoothing method and implement it into future forecasts.
2. The more recent the data, the more heavily weighted it is in the exponential smoothing method. While this method often considers a long period of time in forecasts, it’s smart to weight recent data more heavily because that accounts for current consumer demand. For example, January 2018 wouldn’t be weighted as heavily as January 2021, because many elements have changed in those 3 years, and this method tends to project a similar forecast to that of the previous period. This also means that spikes in data don’t have as much of an effect on the forecast as they do in moving averages.
3. Exponential smoothing produces accurate forecasts. Forecasts produced using this method are accurate and reliable and they predict for the next period. The forecast shows projected demand and actual demand. This allows demand planning to be done effectively, therefore resulting in accurate inventory levels.

## Drawbacks of Exponential Smoothing:

1. The ‘smoothing’ part of this method brushes over high and low variations. As the forecast graph shows a smooth line of data, it’s important to note that spikes in data aren’t necessarily represented.
2. Exponential smoothing doesn’t correctly handle trends. This method is best for short term forecasting, for example the next period. It often predicts future patterns to heavily represent current ones, so it’s not as effective in long term forecasting.

A real-life example can be seen in America’s favorite beverage company, Coca-Cola. Coca-Cola uses a simple exponential smoothing forecast. The stock market is unpredictable and hard to understand for many. Coca-Cola’s investors are often unaware of what the value of their stocks will be tomorrow.

Rather than moving averages with equally weighted values, this works better for them. Exponential smoothing allows Coca-Cola to have a more reliable vision of their near future because they can weight things depending on the age of their prices.

At Avercast, we’ve implemented many different methods into our forecasting software. Of course, these methods include exponential smoothing. If you’d like to benefit from the most accurate forecasting tools available, you’ve come to the right place. With 40 years of supply chain experience, we’ve developed the most reliable forecasting software on the market. Our experts have developed 250+ algorithms to ensure our users get the most accurate and reliable predictions possible.

If you’d like to strengthen your demand planning, you need Avercast. Our team would love to connect with you today. Sign up for a free demo or call to learn more!