Agent-based automation

also known as agentic automation, is the latest technology introduced in the evolution of automation.
Prior to agentic automation, people used machines to implement various forms of automation. However, in recent years, AI technologies such as Large Language Models (LLM), Generative AI, and Large Action Models (LAM) have introduced "agents" that are capable of autonomous decision-making, taking a more active role in performing actions.
For instance, in online customer service, an AI chatbot can communicate with customers, recognize and analyze problems, provide answers, and, when necessary, automatically pass on customer requests to the appropriate department.

The key feature of agentic automation is its ability to recognize and analyze the given environment, think critically, and independently create and execute optimal plans to achieve a set goal.
From the agent’s recognition phase to goal completion, human intervention is minimal, and the agent analyzes relevant data to formulate and implement the best strategy of action.For example, in an automotive production line, robots were previously only involved in the assembly process. With the introduction of agentic automation, however, these robots now monitor the production line, perform assembly tasks, and can automatically reorder parts when inventory is low. They can even adjust production stages to prevent disruptions in the production process.

‍

Differentiating Agentic Automation from Traditional Automation

Robot Process Automation (RPA), which automates repetitive and rule-based tasks using software, remains widely used across industries.
The agent-based automation we’ve discussed, which is based on AI technologies, plays a significantly different role compared to traditional RPA.

RPA (Robotic Process Automation) is highly effective at automating repetitive tasks and can be easily found in e-commerce.
For example, when a customer places an order on an online platform, the RPA in the warehouse can automatically check inventory and prepare the item for shipping.
On the other hand, with the adoption of agentic automation, the system extends beyond simple inventory checks and product deliveries. - it optimizes the entire operational process more efficiently. AI agents can analyze real-time order data to predict demand fluctuations for specific products and automatically maintain optimal inventory levels. Additionally, if a stock shortage is expected to cause shipping delays, the agent can proactively send notifications to customers, enhancing customer satisfaction and improving overall operational efficiency.

The example mentioned above is just one application of agentic automation, but the potential of this technology goes far beyond that.
With agentic automation, we can realize the future of automation in entirely new ways.Complex or unstructured tasks, which were challenging to automate with traditional technologies, can now be easily automated with agentic automation.
For instance, in automobile manufacturing, the customization of assembly according to individual customer orders can be done automatically (with minimal human intervention), leading to the diversification, efficiency, and optimization of the manufacturing process.

‍

In the next post,we will explore the advantages of agentic automation and the changes it will lead to in more detail.