AI and RPA for an automated sales order process

When (conversational) AI meets RPA. 
A collaboration between FLINQ and Arinti to help you automate everyday repetitive tasks.

How many times a day do you have to copy values from an Excel file into a form? How many times a day do you feel like a robot, executing the same task repeatedly?

Take the robot out of the human.

Every job has secondary tasks, which tend to be time-consuming and repetitive. With a secondary task, we mean a task that is not necessarily in your job description but ends up at your desk anyway. Say for example you work at the sales department and part of your job is to process sales orders. Part of this task could be copy-pasting values from PDF files into an ERP system. These kinds of tasks tend to occur multiple times a day and always require the same set of consecutive actions.

Executing the same set of actions repeatedly is exactly what robots are created for. So why wouldn’t you let them help you?

Passing this kind of tasks to a robot, allows you to focus more on the part of your job you like most.

Meet Raphael!

Raphael’s job is to assist employers in the purchase of everyday products for their employees. For example company phones, laptops and many more.

Raphael will ask the employer several questions to get the necessary order details. Artificial Intelligence is used to make this conversation as convenient as possible. For example, authentication happens through facial recognition, Raphael will recommend products based on previous orders and general information like shipping address is stored so it can be reused.

When Raphael has the necessary order details, he will proceed with the ordering of the product. This is where Robotic Process Automation comes into play. As you can see in the video below, Raphael fills in all the details in the SAP system, just as a human would do. Note that for the purpose of this demo we slowed down the RPA part so you can see the different actions of the robot (normally, this process would happen multiple times faster).

Countless possibilities

The use case described in this article is only one of the many use cases in which a combination of Artificial Intelligence and Robotic Process Automation could facilitate tasks. Another problem where AI could assist RPA is in the structuring of unstructured documents. RPA always performs the exact same set of actions, which requires it to receive all the documents in the exact same format. Here Language Understanding can provide a solution. It is possible to train an AI model that understands the different values present in a document and reorganizes the document in such a way that the RPA system understands it.

Creativity and imagination are the only limits to the possibilities and solutions RPA & AI can provide.


On Tuesday, May 21 at the Dynatos Modern Finance Summit, we gave a live demo of Raphael. Below you can find the video of the demo.

Thanks for reading!



Stay tuned.