IBM Digital Business Automation Workshop Exercise - How to combine different automation capabilities
The address information of a corporate customer of a bank have changed. Since the bank is providing a Virtual Assistant service, customer initiates a discussion with the virtual assistant in order to notify the bank of changed address information.
Bank uses AI (IBM Watson Assistant service) to collect the changed address information from the customer in a humanlike dialog. After all needed information has been collected (business id, company name and address information) - virtual assistant initiates a managed business workflow that orchestrates the actual information change into banks CRM system.
Workflow consists of both automated and manual tasks. Robotic Process Automation (RPA) is used to gather the official address information of the customer from the national corporate information web site ytj.fi. All information collected are provided to a human handler in modern UI, where handler can check that the information provided by the customer matches the one got from the ytj.fi web site and then decide to accept or decline the new address information.
- Customer has a discussion with our virtual agent (implemented with Watson Assistant service) to inform the bank of their new address.
- Virtual agent gathers the needed address information and their business id from the customer and then starts a managed workflow (running on IBM Business Automation Workflow environment) to handle this new information.
- First, the workflow triggers a RPA bot (implemented with IBM RPA) that opens the YTJ-site and makes a search using the business id given by the customer. Bot then extracts the official address information from YTJ for that specific business id and sends the results back to the workflow.
- When the bot has gathered the official address information from YTJ, workflow automatically moves forward to its next step. It brings up a human task that one of banks employees need to handle. Workflow shows an UI with both the information customer provided and the information that the bot gathered from YTJ. Now the handler can inspect if these match and can then decide to accept or decline the address information change for the customer.
- The rest of the workflow is fully automated and executes based on the decision handler made. If the change request was accepted, workflow would save new address information to banks CRM system and notify the customer, but if not, just notify the customer of the situation (address information change rejected).
- Blue Demos Environment
- Virtual Assistant
- Connecting Chatbot to Business Automation Workflow
- Automation
- Putting it all together!
If you are running this exercise using yourself reserved IBM Blue Demos environment, you need to complete the LAB 0: Setting Up Your Environment. NOTE! If you are doing this exercise in a class room session with an instructor, you can skip this and go directly to LAB 1. Your instructor will provide separate instructions to access your virtual environment and some of the environment details that you need in LAB 2.
NOTE! You need to be an IBM business partner to access the Blue Demos site and to provision an environment for you.
In this part we create our Virtual Assistant / Chatbot by importing a B2B Banking skill to our Watson Assistant service. We also take a look at how our chatbot has been constructed and we can also test it out.
Here we connect our chatbot to our managed workflow, so that our chatbot can trigger a new managed workflow instance when done collecting information from the banks corporate customer. We will use IBM Cloud Functions (serverless computing services) actions to achieve this.
For this part of the exercise you will be using a virtual environment
running on cloud that you access via your web browser. Your instructor will tell you how to access your environment.
If you manage to finish the labs 1 to 4, you can test and run the hole solution together!
Author: Jukka Juselius (jukka.juselius@fi.ibm.com)
Kudos: Matias Kovero and Sandra Calvo Martinez for helping with the first draft