06/27/2022 | News release | Distributed by Public on 06/27/2022 05:41
by: VMware Senior Member of Technical Staff Peter Amal Raj
Introduction
Addressing customer questions and reducing ad hoc queries is a big challenge for any organization. To mitigate this, organizations are filling FAQ pages and some websites with a lot of details. However, the real challenge is that customers usually do not visit web pages or FAQs to get their answers. The solution VMware IT brings to our platform is the emerging "Chatbot for Business Operations." This is designed and built with end customers and business teams in mind, which provides the respective information in real time and reduces ad hoc requests. Furthermore, it saves customers or business users time by providing the relevant information at fingertips.
The challenge
VMware business teams were spending a lot of time retrieving required information as part of daily operations. The various information was hidden across various application pages and the business teams were spending valuable time navigating around this.
Due to this, business operations and decision-making time were being delayed, contrary to our vision of, providing business information to users' fingertips with less effort and in less time.
To fulfil this vision, VMware IT built a chatbot that works with our artificial intelligence (AI) and microservice architecture to ease business operations.
The implementation
This is an interactive-based chatbot that helps business users retrieve their required business data in practically no time at all. The best part about this chatbot is that business users do not need special training to use it. It can be integrated with any UI channel.
The Chatbot flow starts with the business user command/request from the UI channel. The user request will be received and passed on to the processing manager. The processing manager will feed the request into the AI system to convert the user request to the machine-understandable format (intents and entities). One of the recommended open-sources AI is "RASA AI"
AI system will send back the intent and entity details back to the processing manager. Based on the values, the processing manager will call the authentication service to validate the user permission. If the authentication is passed, the respective microservice end points will be triggered to get the business data. Once the data is retrieved, based on the UI channel, the UI response will be formatted and displayed back to the business users. See Figure 1.
Figure1: Chatbot architecture
For the Chatbot, role-based authentication system is the right choice as it provides a deep level of control for user authentication.
Some implemented business scenarios
Advantages
What's next?
Currently, the chatbot has been implemented to connect only to the Oracle EBS Order Management System. The upcoming plan is to extend the services to add SAP-related business scenarios.
We're also planning to expand the usage to various business teams, including Sales, that will reduce their workload tickets.
VMware on VMware blogs are written by IT subject matter experts sharing stories about our digital transformation using VMware products and services in a global production environment. Contact your sales rep or [email protected] to schedule a briefing on this topic. Visit the VMware on VMware microsite and follow us on Twitter.