Intelligent Process Automation
Intelligent Process Automation (IPA) utilises existing RPA technologies alongside AI capabilities in order to automate and improve business processes. Where RPA aims to replicate human actions upon interfaces in order to improve scalability and reduce costs, IPA sees the robot handling all aspects of a process, improving and eventually surpassing human capability.
McKinsey describes IPA through five constituent components: RPA, smart workflow, machine learning, natural language generation (NLG) and cognitive agents. The latter three in particular broaden the scope for business processes automated. NLG allows for robot interpretation of data past strict RPA rules; one application of this is to create more advanced forms of text based documentation. Machine learning allows for continual improvements of the process and the ability to make decisions as a human might. Cognitive agents could then use both of these tools to perform and communicate tasks usually reserved for human judgement.
This leap from copying dictated human actions to predicting and implementing real human behaviour allows for more powerful and dynamic automation solutions. The ability to find meaning in data allows the robot to complete complex processes from start to completion without human intervention. The addition of human-like decision making also makes process automation a more viable solution for front office tasks.
IPA can be seen as an extension of RPA, utilising these emerging technologies to augment automation in the limiting areas of established RPA. Whilst current RPA tools are excellent for automating a large variety of logic-based processes, the future of AI technology could open up the possibilities for what automation is capable of.
All views and positions contained herein are entirely my own and not representative of any other individual, organisation or company.