Annons

How Do You Escape the Proof-of-Concept Purgatory and Translate AI Into Measurable Results?

IFS EVP MARTIN GUNNARSSON on scaling up and translating AI into real results.
Agentic AI is one of the concepts that we are increasingly seeing as AI develops. There are several definitions, but an exemplary simple and illustrative one is that agentic AI can be seen as a framework; while AI agents are the building blocks within the framework. But a little more concrete: What can the agents, the building blocks, do in concrete terms?
Business system developer IFS is one of the developers that has quickly come a long way in the area related to industrial AI, and in today's guest column, the company's Martin Gunnarsson, senior VP for product and partner strategy, discusses more around the matter.
IFS has developed an interesting model with its sights set on agentic AI, where AI agents play important roles in the setup. These are programs that can interact with their environment, collect data and use the data to perform self-determined tasks to meet predetermined goals. Some agents work only with predefined rules, while others use learning algorithms to refine their behavior. Humans set goals, but an AI agent independently chooses the best actions it needs to take to achieve those goals. They can do what needs to be done better, faster and, not least, cheaper than things have been done before. Using sophisticated models, AI agents can, for example, infer a customer's intent, predict needs and offer tailored solutions, all while working 24/7 to ensure consistent and effective support. In this, Gunnarsson and IFS see the emergence of AI agents that can handle complete business tasks across different parts of an enterprise system. We can thus move from simple automation to complex, context-aware actions. IFS systems should be a bridge between all these islands of possibilities. Martin Gunnarsson writes:
"Many industrial companies have already started to explore the possibilities of using agentic AI where intelligent agents can independently take over tasks that one or more people currently perform. For example, an AI agent can monitor sales orders, warn of missing items and suggest replacements. In parallel, another AI agent can write a customer email, which you review and confirm before sending. But despite a strong desire to utilize AI technology, many are still in a pilot phase, stuck in a kind of "Proof-of-Concept" purgatory - in which AI is often more of a burden than a real asset. How do you take the step to scale up AI so that it can seriously streamline operations?

Without clear business goals – no measurable results
The first step is to clearly define which business goals the AI ​​agents should achieve. An AI agent should have a clear mission, for example a ”monitoring agent” should monitor and improve the uptime of a machine, the “communication agent” focuses on communicating that there is a problem with the uptime of a particular machine. By linking specific roles with clear goals for each agent, measurable results are created

The next step is to ensure that the data and infrastructure are ready. The success of AI agents depends on access to high-quality data and robust processes. Improving data infrastructure and creating clear guidelines for how the business should function is crucial to avoid AI projects stopping at the prototype stage.
It is also important to have a phased implementation; starting by implementing the use of AI in the most important parts of the business. It may sound obvious, but too often companies are tempted to try to run a full-scale launch right away.

Counteract staff fear of AI
For AI to become an integral part of the company, of course, the organization and staff must also be on board. A major challenge today is that many employees worry that AI will threaten their jobs. In this case, It means creating clear plans for the transition, educating staff and involving them early on – so that they understand that AI should be seen as a tool to free up time for more value-creating work, not as a threat. It’s about the company’s competitiveness over time, where AI will play a crucial role in the company’s future growth and profitability

Finally, continuous measurement and improvement are extremely important. Evaluating how AI agents perform and gradually giving them more autonomy is the key to scaling up. At the same time, the organization must have clear frameworks to guide development and ensure that AI acts in line with the company’s values ​​and goals.
I am convinced that the companies that take these steps on their AI journey will be the market winners. Because the industrial AI revolution is about translating the enormous opportunities we now face into real results.

By Martin Gunnarsson, SVP Product & Partner Strategy, IFS

Print Friendly, PDF & Email

Success Stories

Success Stories

Industriellt

Intressant på PLM TV News

PLM TV News

PLM TV News

PLM TV News

PLM TV News

PLM TV News

Aktuell ANALYS

Aktuell Analys

Aktuell Analys

3D-printing

Block title