It’s been a while since we did a Team Picks post, but you can bet everyone’s been busy exploring SAP Business One version 9.2’s awesome new features!
You wouldn’t plan a day to the beach without checking the weather, and you shouldn’t perform MRP (Materials Resources Planning) without proper forecasting ahead of time. That’s where the handy feature “Intelligent Forecast” comes into play.
SAP Business One, particularly SAP Business One version for SAP HANA, is packed full of tools for storing and interpreting data. You may already be using the basic MRP forecasting method available in older versions of SAP Business One – “Intelligent Forecast” is an enhanced version of the feature. It is currently only available in SAP Business One Version 9.2 for SAP HANA.
Intelligent Forecast Features
- Hindcast to dynamically adjust outlier and what-if scenarios.
- Forecast results can be used in MRP wizard.
- Easily forecast and leverage what-if-analysis to improve decision-making.
The Intelligent Forecast is a statistical forecast tool with built-in models incorporating trends and seasonal factors. It includes 2 forecasting methods. SAP Business One automatically selects the best algorithm.
TESM stands for Triple Exponential Smoothing. TESM is used to handle time series data containing seasonal components. It works by incorporating a stationary component, trends, and seasonal factors. Both the trend and seasonal factors can be additive or multiplicative in nature.
For example, you may sell 200 more fruit baskets in April than March, in which case this trend is additive in nature. On the other hand, you may sell 500% more fruit baskets in February (thanks to the Chinese New Year) and December (thanks to the Christmas season) than other months in the year, in which case your seasonal factor is multiplicative in nature.
LRDTSA stands for Linear Regression with Damped Trend and Seasonal Adjust. It is chosen for forecasting when times series data presents a trend. A damped smoothing parameter is used to smooth forecasted values and prevent over-casting. This method also detects seasonality in your data in order to adjust your forecasting results.
The Purpose of Intelligent Forecast(ing)
The Intelligent Forecast tool aims to help you make better predictions, and consequently more well-informed business decisions. SAP Business One uses statistical methods to analyse your past data and provide estimations for demand based on historical trends.
For example, seasonal peaks which recur annually could lead the forecast to suggest higher demand for these periods. You can ensure that you maximize your sales potential during these seasons by producing or purchasing more stock to avoid stockouts, and driving higher marketing activities for these periods ahead of time, instead of reacting to them as and when they occur.
Steps to perform Intelligent Forecasting
1. Decide your goals/purpose of generating the forecast
Perhaps you want to predict market demand for Product A within the next 6 months, and carry out MRP (Materials Resource Planning) based on this.
2. Create a new forecast
Head to the Intelligent Forecast tool by navigating to MRP → Forecasts.
Enter a new Forecast Code and Forecast Name for a specific time period.
For example, you may want to name the forecast for the first half of this year 2016 1H.
Set the Start Date and End Date for your forecast – e.g. 1 February 2016 to 31 July 2016 for 2016 1H.
Choose whether you want to view daily, weekly, or monthly predictions.
Click on Generate Forecast , and select Intelligent Forecast from the dropdown menu.
(Please note that his screenshot was cropped for visual purposes. The actual window is wider.)
In the Generate Intelligent Forecast window, you may choose to select items by item, preferred supplier, or default warehouse.
Select your item range by clicking the item number fields and selecting your chosen items.
Choose whether to generate the forecast based on sales history by sales order, delivery documents, or A/R invoices.
Click on forecast button on the right side of your window beside these options.
3. Adjust your Intelligent Forecast
The generated intelligent forecast shows predicted demand for your selected products per warehouse and per day/week/month, depending on which view you selected earlier.
The blue lines in your forecast chart show your historical sales data based on sales orders/deliveries/A/R invoice, depending on your selection earlier.
The orange lines show your predicted demand for the time period requested.
If you spot an outlier in your past data, such as a large one-off deal which is unlikely to happen again, it may affect your prediction results and skew it such that accuracy is compromised.
Click and hold on to the outlying point and drag it down or up to a more reasonable figure, depending on whether it is uncharacteristically high or low.
The change in your prediction results will be shown immediately.
3.2 Change your historical data period
You may also adjust the number of maximum history time buckets used to generate your forecast, e.g. the last 100 days, weeks, or months, depending on which measurement/unit of time you used earlier.
Maybe you just started your business 2 years ago, and the first year’s data may no longer be relevant as your sales volume has picked up significantly since then. Or perhaps the economy has slowed down recently, and you want to view a more conservative forecast by only analysing the past year’s data.
To estimate how reliable your prediction is, use the hindcasting function.
SAP Business One predicts demand based on a date in the past – look for the ↤ symbol (on the left in the screenshot above) and drag it to a suitable date in the past by looking at the shape of your graph.
Instead of generating your forecast from today’s date, the forecast is generated from the past date you have just selected.
For example, if you drag the blue bar to 15 November 2015, your forecast will now be generated from 15 November 2015 to 31 July 2016, instead of starting from 1 February 2016.
You can compare how accurate or inaccurate it is in comparison to the actual results shown.
If the values are not too different, you can be fairly sure that the forecast is accurate and reliable.
Click Save and Close on the bottom left of your screen.
The Generate Intelligent Forecast window will close.
Click Add in your Forecasts window. Your new Intelligent Forecast will be saved in your MRP process for future use.
And there you have it! You’ve just performed Intelligent Forecasting in SAP Business One.
Here’s a video walkthrough of the steps, if you prefer:
Intelligent Forecast vs Basic Forecast
If you haven’t used the Basic Forecast feature in SAP Business One, you may be wondering what the difference is between the Basic Forecast and Intelligent Forecast.
The Basic Forecast uses different algorithms from Intelligent Forecast to generate predictions. There are four forecasting methods in the Basic Forecast:
1. Simple Average
The total quantities of each selected item are extracted from the sales documents you have selected (Sales Orders, Deliveries, A/R Invoices) within the history start date and end date (similar to your Time Buckets in Intelligent Forecast) and summed up.
The summed quantities are then averaged by the number of days, weeks, or months for your forecasted period.
2. Daily Average
The total quantities of each selected item are extracted from the sales documents you have selected for each day within the history start date and end date and populated into each day within your forecast start and end dates.
3. Weekly Average
The total quantities of each selected item are extracted from the sales documents you have selected for each week and summed up. The summed quantities per week are populated into each week within your forecast start and end dates.
4. Monthly Average
The total quantities of each selected item are extracted from the sales documents you have selected for each month are summed up. The summed quantities per month are populated into each month within your forecast start and end dates.
The hindcast feature in the Intelligent Forecast is not available in the Basic Forecast.
In the Intelligent Forecast, you may drag data points up or down to adjust for outliers, giving you greater control over your forecast outputs.
In Basic Forecast, you can rely on Forecast Quantity Adjustment to adjust your forecast outcomes as a whole, by increasing or decreasing target quantities by a certain percentage. This method is less precise, since it affects all quantities projected for, instead of correcting only one or a few specific points.