Skip to Content
Product Information
Author's profile photo Sai Giridhar Kasturi

Predicting delivery delays with Machine Learning – (Part 2)

Introduction

This is the continuation of the blog Predicting delivery delays with Machine Learning – (Part 1)

Predicting delivery delays with Machine Learning – (Part 1)

Solution Model

The ‘Predicted Delivery Delay’ app is built by using Analytical List Page (ALP). This app allows to examine the business data from various perspectives, investigate a root cause for a specific issue using the drill downs and provide a detailed analysis of planned, actual and predictive figures. The tile of the app provides a quick view of the number of open order items that are late, early and on-time. This app is accessible on the Fiori launch pad under the business catalog ‘Sales Order Monitoring’.

  • Business role: SAP_BR_INTERNAL_SALES_REP (Internal Sales Representative)
  • Business catalog: SAP_SD_BC_DELIV_DLY_PRDT (Sales – Delivery Delay Prediction)

8.png

Below is an overview page with header (red) and content area (green) of the predicted delivery delay app.

§ Header area provides a visual representation of the data.

§ Content area provides the detailed values of the key parameter selected in the header area.

With this app, you can display the planned delivery creation date and predicted delay for sales document items. Kindly note that the planned delivery creation date is only calculated for the confirmed schedule lines.

9.png

The header area can be viewed in two types of filters – Compact filter and Visual filter.

Compact filter – this is traditional display of filter values to get the desired result. You can use the compact filter with default parameters – sales document, sold-to party, sales organization, material, and plant. If necessary, you may choose to add additional filters (27 more filter options available in this case) such as division, distribution channel, route, plant, MRP area etc. Fields, sales document and sold-to party are allowed for fuzzy search with this filter option.

10.png

11.png

The other display option is the visual filter, which is a graphical representation of filter values in chart form.

12.png

The visual filter is available with 3 possibilities. Line Chart, Donut Chart and Bar chart. In case of visual display, the filter options are limited to “Number of Items by Sold-To Party” and “Number of Items by Plant”. Provision of additional filters options is currently not available.

With adapt filter option in the header, it is possible to customize visual display area with additional features. Users can navigate to choose the appropriate chart type (line / donut / bar) or the value measure (No. of items, predicted delay of delivery creation, requested quantity, cumulative confirmed quantity, net value) or the sort sequence as per the business requirement.

Kindly note that the parameters selected in the compact filter (for sales group, sales office, sales organization, distribution channel, division, and sold-to party) will influence what is displayed in the visual filter section. For example, if you select a specific sold-to party or a plant in the compact filter, then the visual filters display the results for this sold-to party only.

Similarly, for selections made for one filter in the visual filters influence the other. That is if you select a sold-to party in one visual filter, the second visual filter is reloaded and filtered on this sold-to party, and vice versa. The chosen filter values can be saved as a variant so that the selected layout can be easily accessed next time.

14.png

The content area reflects the data based on the selected parameter from the header area. It can be displayed in 3 different views

1. Chart and Table view (default)

2. Chart view

3. Table view

15.png

16.png

17.png

Further, within the content screen the users can choose to display the data in different types of chart options available such as vertical or horizontal bar chart, pie chart, line chart, heat map etc.

18.png

Also, the table view format allows to export the content details into an excel format.

19.png

The “view by” option available for chart display, allows to choose additional dimensions to the content area. For example, additional dimensions can be added to the “Total Predicted Delivery Delay Status” such as Sales Org, Division, Material Group etc. for example.

20.png

Further, hovering/selecting a specific item in the chart area can display a detailed summary as shown below.

21.png

In case of table view, the content area can be maximized if required with a flexibility to move the columns and sort the values with a different parameter.

22.png

23.png

Further, in addition to predicting the delivery delays, this app provides access to other relevant apps to resolve delivery issues (if exist). A hyperlink to sales document number in the content area allows to navigate to the apps as listed in the snapshot below.

For example, using the Track Sales Order Details app you can view the details of delivery issues, considering the predicted delivery delay column and understand why a specific transaction is delayed or delivered early.

24.png

Highlight the line item (sales order item) and click on ‘Details’ to see the end-to-end document flow and status of each document in the transaction. The details are displayed from the ‘Track Sales Order’ app.

25.png

26.png

Conclusion

Similar to other two machine learning scenarios covered in my previous blogs, this embedded analytical app is easy to navigate, compatible with desktop / tablet devices and offers various functions to address day-to-day operational issues. In addition to predicting the delay in delivery, this app’s ability to navigate to each sales document in detail, can be a true value to business.

Reference Link:

Predicting delivery delays with Machine Learning – (Part 1)

Assigned Tags

      5 Comments
      You must be Logged on to comment or reply to a post.
      Author's profile photo Priyank Kumar Jain
      Priyank Kumar Jain

      Nice one Sai !!

      Author's profile photo Paul Taylor
      Paul Taylor

      Great follow up from part 1 !

      Author's profile photo Prateek Agarwal
      Prateek Agarwal

      Very well explained.

      Author's profile photo Mateus Blume
      Mateus Blume

      Great explanation using many pictures, thank you!

      Author's profile photo Nicolas Van der Laan
      Nicolas Van der Laan

      Thank you for the explanations. I have tried to use this application but I am encountering a problem. In my dataset I have a lot of deliveries that are created late and goods issues as well. However in the forecast application none of my future orders are announced late. When I saw my dataset I expected to have some late orders. Do you have any answers?