Introducing of RPA classes


1. RPA Class 1(an Automation of routine work)

2. RPA Class 2 (an Automation of some irregular work)

3. RPA class 3(a highly autonomous)


1. RPA Class 1(an Automation of routine work)

rpa class
rpa class

RPA class 1 (Class 1) basically means automating the PC work performed by humans. This includes “information acquisition” (Excel etc.) “reading information (structured data)” “input work” “verification work” “login to multiple systems”. Information acquisition means automatically acquiring data from Excel and creating contracts and reports. Reading information means reading traditional structured database information (customer data, numerical data, etc.).

Input work is to automatically enter the acquired data. The verification work verifies that the data is correct, as was the case with benefits. Logging in to multiple systems means that the bot automatically sends the ID and password to other systems to log in to the system automatically.


2. RPA Class 2 (an Automation of some irregular work)

RPA Class 2 refers to automation of exception handling and atypical work. Class 2 is also known as “Cognitive AI” Exception handling is learning based on data analysis using AI (deep learning). Atypical work refers to the processing of natural language (unstructured data). For example, artificial intelligence is software that enables computers to understand, inferences, and learn on behalf of humans, rather than letting computers automate certain processes. A typical task is the processing of unstructured data (such as audio and images). The Class 2 is also applicable to SMEs who want to digitize conventional operations and promote DX.


3. RPA class 3(a highly autonomous)

RPA Class 3 not only automates work with advanced artificial intelligence, but also performs “decision making” “complex processing” “advanced analytics”.The technologies required for “high degree of autonomy” are as follows

  • Natural language processing

Natural language processing is the processing and processing of “natural language” that humans usually speak and write using AI.

  • Big data analytics

Big data analysis is, for example, analysis of customer information and purchasing information of large companies, which is useful for sales and sales activities.

  • Artificial intelligence (AI)

Artificial intelligence is software that allows computers to understand, inferences, and learn on behalf of humans.

  • Machine learning

Machine learning is a computer system that uses certain patterns and inferences to efficiently carry out specific tasks.

  • Large-scale processing

Large-scale processing is similar to processing “giant macros” as in Class 1. However, unlike Class 1, Class 3 differs in that it processes large amounts of information that is constantly changing by AI.

  • Autonomous adaptation

Autonomous adaptation includes, for example, the current topic of “autonomous driving” technology. Using 5G radio waves and GPS signals, the EV connects to AI and the cloud, and the EV calculates the route itself and carries the passengers to the destination. In addition, AI constantly learns user’s favorite music and comfortable air conditioning temperature in the car. Providing a comfortable environment for users is autonomous adaptation.

In other words, AI was used to understand and process natural language that class 2 cognitive AI could not process in the past. However, in the highly autonomous world of Class 3, AI itself analyzes and processes information, and even makes decisions. The Class 3 includes major companies, research institutes, and government agencies that introduce such large-scale systems.

Reference URL: