Glossary: What does AI mean? 

AI for business process optimization is at the heart of what we do. Here we explain the technologies that are essential to us and how we apply them to our software solutions!

C

Copilot

is a digital, AI-based assistant that actively supports users in the operation of software. Using machine learning and large language models, a copilot understands natural language, recognizes user intentions and directly provides suitable recommendations or automations for next steps. Copilots continuously learn from interactions with the user and thus become better and better at making complex tasks simple and easy to understand.

Application with INFORM

INFORM specifically integrates co-pilots into software solutions. These AI helpers enable operation by voice input and ensure that even complex processes can be controlled intuitively - regardless of the user's level of experience. A chatbot can also be found on the homepage of our software. It answers all questions about INFORM without visitors having to click through the website.

D

Data Mining

refers to a computer-aided process for identifying hidden patterns, correlations and trends in large amounts of data. It combines statistical analysis, machine learning and database systems to gain usable insights from heterogeneous data sources. The aim is not only to analyze, but also to usefully consolidate information for well-founded decisions, strategic planning and predictive analyses. Data mining is one of the important elements that INFORM summarizes under the term Process AI.

Application with INFORM

INFORM uses data mining to gain data-based insights that flow into the calculation of software solutions for optimizing warehouse, logistics or resource scenarios, for example. Our software for supply chain management uses data mining to identify sales trends, for example. For example, manufacturers can use historical sales figures and market trends to identify at an early stage which product variants will be particularly popular in the future or find out which products are typically bought together.  This makes it easier to plan stock levels and optimize production and replenishment processes.

Decision Intelligence (DI)

supports companies in making operational decisions faster, more consistently and more cost-efficiently. It combines digital analysis capability with decision-making logic, enabling options for action that can be implemented immediately and create real added value. In complex and dynamic environments - such as time pressure, limited resources or changing priorities - our software systems intelligently correlate a wide range of factors on the basis of DI. The result is a holistic basis for decision-making that keeps an eye on the overall operational optimum.

Application with INFORM

INFORM has been using Decision Intelligence as a core function of its software solutions for over 50 years. The ability of our software to relate a wide variety of factors, constraints and prioritizations to each other in a meaningful way enables a holistic solution that always serves the overall optimum of an operational process. This development is a pioneering achievement by INFORM.

F

Foundation Models

are comprehensive AI models that have been pre-trained on large, diverse data sets and can be used for specific applications through fine-tuning. Foundation models build on existing knowledge so that a new AI model does not have to be developed each time. Examples of this include large language models that can understand and generate speech.

Application with INFORM

INFORM uses foundation models as a flexible basis to provide specific AI applications within its software solutions efficiently and quickly, thus enabling customized optimizations for customer requirements. This includes time series analyses, for example. In the area of supply chain management, they are used for sales planning, among other things. Here, historical data is used to calculate future-relevant patterns that provide information about the future sales of a product.

L

Large Language Models (LLM)

are highly developed AI models that can analyze, understand and generate large quantities of natural language. They are based on so-called transformers, a subclass of neural networks. They are trained with huge amounts of text data, enabling LLMs to capture semantic relationships and answer complex questions. Typical applications include chatbots, automatic text generation or supporting users in the operation of complex systems using natural language. They can also make existing specialist knowledge available across departmental and system boundaries as part of enterprise knowledge management. To this end, they extract and analyze information from unstructured data such as emails, PDF and Word files, contracts or invoices and make it available to users on request in a tailored manner.

Application with INFORM

INFORM integrates LLMs into its software solutions to enable users to interact easily and naturally - for example through voice-controlled digital assistants and co-pilots.

M

Machine learning

is an area of artificial intelligence that focuses on algorithms that learn from data and independently derive predictions or decisions from it. The aim is to recognize patterns, understand correlations and draw conclusions about future situations from known data. A distinction is made between supervised learning, unsupervised learning and reinforcement learning. These methods use neural networks and decision trees, among other things. Machine learning enables progress in areas such as speech recognition, image analysis and predictive modeling.

Application with INFORM

The results of machine learning provide a precise basis for our optimization algorithms. This also makes the "optimal result" that our software calculates even more accurate - especially where dynamic adaptation to regularly changing conditions is required, such as in production planning, fraud detection or workforce scheduling

N

Neural networks

are a central component of AI. Inspired by the human brain, they consist of networked, adaptive, artificial neurons. They can recognize complex patterns, interpret information and make decisions independently. Due to their high learning capacity, neural networks are particularly suitable for complex tasks such as image recognition, speech processing or predictions based on large amounts of data.

Application with INFORM

INFORM uses neural networks in its software solutions to solve challenging optimization problems and to incorporate information that cannot be efficiently evaluated by conventional algorithms.

O

Operations Research (OR)

is a scientific method for the systematic, mathematical analysis and optimization of complex decision-making processes. Mathematical models and algorithms are used to calculate optimal solutions for planning and control problems. The areas of application range from route planning and determining the optimal use of resources to the real-time control of time-critical processes.

Application with INFORM

INFORM uses operations research as the central technology of all software solutions. OR methods form the basis for the optimal calculation and control of complex business processes, for example from logistics to production and retail.

P

Process AI

refers to the use of artificial intelligence to optimize business processes. At INFORM, we are expanding our proven decision-intelligent solutions with learning, contextualizing and generative AI functionalities. Process AI uses these technologies to intelligently control operational and tactical processes - for example in production planning, transport logistics or workforce scheduling. Large process models are used to learn, digitally represent, analyze and improve the conditions and relationships of process landscapes.

The vision of INFORM:

An AI that thinks along with people, that fully understands entire process worlds and that integrates closely with the real working world of companies. This is what we are researching at INFORM.
 

Process Mining

is a method of making business processes visible and traceable in real time during their daily execution. For this purpose, objects are tracked as they pass through a process - such as a suitcase during baggage handling at the airport or a component as part of a production process.  Using digital event data, process mining reconstructs the actual processes in the company and uncovers deviations between planned and actual process sequences. This allows bottlenecks, inefficiencies and potential for improvement to be identified in a targeted manner.

Application with INFORM

INFORM uses process mining to analyze processes in detail and thus gain valuable insights - for example in ground handling at airports or in logistics. Our software reveals where processes in the company are not running smoothly and helps to optimize them in a targeted manner.