Robotic Process Automation

In my previous blog, I spoke about the scalability and throughput of the platform. Manual processes have the same problem and most are mitigated with traditional software and/or with integration across solutions. However there are several scenarios, particularly cross organisation data flow like purchase order (PO) or invoice processing, where traditional software does not work and integration across two companies doesn’t exist. For these reasons these processes are largely manually and robotic process automation is the only reasonable choice.

Let’s consider the scenario where a customer places an order with their vendor through a purchase order. The customer creates the PO in their system and faxes / emails it to the vendor. The recipient at the vendor receives this PO and inputs it into their software letting the service or product delivery teams know that there is a request for their offering. This does not scale as the number of PO increases as in the case of one of our customers.

Our client is a large manufacturer in the power solutions industry. One of their customers implemented just-in-time manufacturing (JIT). JIT is a methodology that ensures the right amount of inventory for both raw materials and finished goods. In order to ensure this their system automatically generates POs as and when the inventory level of the raw materials drop below the defined threshold. This resulted in our customer receiving 30+ POs in a day and manually updating their SAP system was cumbersome and not happening on time.

We implemented our Robotic Process Automation (RPA) solution for the client.

The steps involved were as follows

1. The POs were automatically sent to the Smart Platform by the client.
2. Smart Platform validated the message for spam / virus / junk and only those messages that passed the check were forwarded to the client’s application.
3. The email with the PO is processed.
4. POs having errors like price difference etc. are flagged and the relevant user are automatically notified.
5. A python bot is configured at the client’s end. The bot periodically logs in to Smart Platform.
6. Bot downloads the unprocessed PO as CSV
7. Bot then logs in to SAP system and uploads the CSV and creates POs
8. The POs are then flagged as processed on Smart Platform.

With this implementation of the RPA solution, the client was able to eliminate the manual steps and process the POs with no errors and within their defined turn-around-times enabling them to deliver their orders on time.

Smart Platform was extended with the capability to receive and process incoming emails, Bot library with reusable components and a Bot client customised for the clients process and systems.